Final Thesis-Print Out Taken-2!4!14

FINITE ELEMENT ANALYSIS OF FUNCTIONALLY GRADED
PLATES AND SHELLS
Ph.D. THESIS
by
GULSHAN TAJ M.N.A
DEPARTMENT OF CIVIL ENGINEERING

INDIAN INSTITUTE OF TECHNOLOGY ROORKEE
ROORKEE-247667, UTTARAKHAND, INDIA
APRIL, 2014
FINITE ELEMENT ANALYSIS OF FUNCTIONALLY GRADED
PLATES AND SHELLS
A THESIS
Submitted in partial fulfilment of the
requirements for the award of the degree
of
DOCTOR OF PHILOSOPHY
in
CIVIL ENGINEERING
by
GULSHAN TAJ M.N.A
DEPARTMENT OF CIVIL ENGINEERING

ROORKEE-247667, UTTARAKHAND, INDIA
APRIL, 2014
©INDIAN INSTITUTE OF TECHNOLOGY ROORKEE, ROORKEE – 2014
ALL RIGHTS RESERVED
ROORKEE
CANDIDATE’S DECLARATION
I hereby certify that the work which is being presented in this thesis entitled “FINITE
ELEMENT ANALYSIS OF FUNCTIONALLY GRADED PLATES AND SHELLS” in
partial fulfilment of the requirements for the award of the Degree of Doctor of
Philosophy and submitted in the Department of Civil Engineering, Indian Institute of
Technology Roorkee, Roorkee is an authentic record of my own work carried out at
Department of Civil Engineering during the period from July, 2011 to April, 2014
under the supervision of Dr. Anupam Chakrabarti, Associate Professor,
Department of Civil Engineering, Indian Institute of Technology Roorkee, Roorkee,
India.
The matter presented in this thesis has not been submitted by me for the
award of any other degree of this or any other Institute.
(Gulshan Taj M.N.A)
This is to certify that the above statement made by the candidate is correct to
the best of my knowledge.
( Anupam Chakrabarti)
Supervisor
Date: /04/2014
The Ph.D. Viva-Voce Examination of Ms. Gulshan Taj M.N.A, Research

Scholar, has been held on _____________________________________________.
Supervisor Chairman, SRC External Examiner
Head of the Department/Chairman,ODC

ABSTRACT
Functionally graded materials (FGMs) are advanced class of engineering composites

constituted of two or more distinct phase materials described by continuous and smooth
varying composition of material properties in the required direction. The mechanical
properties such as Young’s modulus, Poisson’s ratio, shear modulus, and density are varied
according to simple rule of mixture in terms of volume fraction distribution of constituents.
Potential reduction of in-plane and transverse through-the-thickness stresses, reduced
stress intensity factors, an improved residual stress distribution, and higher fracture
toughness are few of the advantages offered by these advanced composites. In addition,
these materials are capable of surviving high temperatures and large temperature gradients
that may occur within a fraction of seconds in some structures such as aircraft, while
preserving their structural integrity. Thin walled rotating blades that could be used in
helicopters and turbo machinery fields, thermal barrier coatings, nuclear reactors, micro
and nano devices, dental and medical implants, and piezoelectric and thermo electric
devices are some of the areas where the concept of FGM has been successfully
implemented in the modern era.
The proper choice of homogenization approach used in FGM structures should be
based on the gradient of gradation relative to the size of a representative volume element
(RVE). In literatures, some of the averaging methods that incorporate the homogenous
nature at RVE scale and heterogeneous nature at global scale are proposed. Among
different methods, rule of mixture is widely employed in many studies, while few assume
the Mori-Tanaka and self consistent schemes. Mori-Tanaka based homogenization
approach accounts for the interactions among the adjacent inclusions. In the present study,
both simple rule of mixture and Mori-Tanaka scheme are employed to homogenize certain
mechanical/thermal properties of the constituents.
In highly heterogeneous structures like FGMs, membrane-flexure coupling exist
due to anti-symmetric nature of material properties. Hence, it seems important to consider
the improved structural kinematics in the form of accurate variation of in-plane and
transverse displacement components that describe the realistic parabolic distribution of
transverse shear deformation. In the present study static, free/forced vibration and buckling
responses of FGM skew plate/shell structures are performed by developing an efficient
displacement based 2D FE model. A higher order shear deformation theory (HSDT) which
i
accounts for realistic parabolic variation of transverse shear deformation is employed in
the present thesis work for this purpose. In case of thermal analysis, one-dimensional
Fourier heat conduction equation has been solved by imposing appropriate thermal
boundary conditions at the top and bottom surfaces. Temperature dependent material
properties are also incorporated in the analyses that follow the simple rule of mixture of
materials.
During the implementation of higher order theory in finite element (FE) method,
the problem of C1 continuity is encountered due to the existence of first order derivatives
of transverse displacement component in the expression of in-plane fields. In the present
FE formulation, this problem has been circumvented by means of appropriate substitution
of independent nodal unknowns and thus leading to an efficient C0 FE formulation. A nine
node isoparametric Lagrangian element is used to mesh the assumed plate/shell geometry.
In case of skew boundary, for nodes lying on the skew edges, suitable transformation rule
has been employed to generate the corresponding global element matrices. While
incorporating the strain field for FGM shell structures, all three curvature terms i.e., 1/Rx,
1/Ry and 1/Rxy are included for the analysis of different shell forms like hyperbolic
paraboloid and hypar shells which are not included in any other studies so far.
The conventional power law distribution adopted in most of the literatures leads to
the configuration where the top layer will be ceramic rich, and bottom layer will be rich in
metal with composite structures in between. But, due to some practical requirements, the
design may demand for probable combinations of FGM configuration other than
conventional one. To meet this criterion, FGM plate described by a four-parameter power
function is incorporated in the present work for static, free vibration and buckling
problems. Four-parameters that define the distribution law are suitably assumed to achieve
the conventional, symmetric and asymmetric FGM profiles by satisfying the volumetric
relationships between the constituents.
Due to the large jump of material properties at the layer interfaces, the problem of
delamination is generally observed in conventional sandwich structures. The concept of
FGM is also employed in such sandwich structures to overcome this problem. In the
present research work, an improved structural kinematics that account for realistic
variation of transverse shear component and quadratic thickness variation of transverse
displacement was employed for static and free vibration analyses of FGM skew sandwich
plates/shells. FGM skew sandwich plate/shells are modeled by two kinds of approaches. In
the first case, the core part is designated by ceramic component; while top and bottom
ii
layers are rich in metal component thus leading to an FGM profile at the top and bottom
layers. In the later case, the top and bottom layers are defined by ceramic and metal
components and hence the core portion is defined by functional grading. By designating
the bottom-core-top layer thickness of the plate/shell with respect to overall thickness,
different kinds of sandwich plates/shells are modeled in the present study.
During their service life, structures may undergo large deformation under
mechanical, thermal and thermo-mechanical loading conditions. In such cases, the
equilibrium equations should be written in terms of the deformed configuration of the
structures by non-linear strain-displacement relationships. Hence, to analyze FGM
sandwich structures under large deformation, a non-linear FE formulation is implemented
in the frame work of Green-Lagrange form of equations. In this work, FGM skew
plates/shells constituted by single/multi layers are analyzed under large deformation.
Newton-Raphson iteration scheme is employed to trace the load-displacement path. When
in-plane loads are high in magnitude it is necessary to consider the secondary bifurcation
stage which is known as post buckling stage of structures. To trace the post buckling
equilibrium path of FGM sandwich plates, a direct iterative algorithm is employed to get
the converged frequency values under different magnitude of amplitude ratios.
Many new results based on various geometric properties such as aspect ratio,
thickness ratio, curvature-side ratio, boundary conditions and material properties like
volume fraction index and material constituents are accomplished in the present research
work to perform linear/non-linear static, free/forced vibration, and buckling analyses of
FGM skew plates/shells constituted of single and multiple layers. The various imperative
conclusions arrived from the present research work should be useful for researchers,
analysts and designers engaged in this area.
iii
ACKNOWLEDGEMENT
I wish to express my sincerest gratitude to my supervisor Dr. Anupam Chakrabarti
(Associate Professor), Department of Civil Engineering, Indian Institute of Technology,
Roorkee for his consistent advice, guidance and encouragement throughout the period of
this research work right from the inception of the problem to the final preparation of the
manuscript. He has shown keen interest in and offered valuable suggestions on even the
minute details of my research work throughout the research period. His deep understanding
of the relevant topics as well as grip over computer programming, constructive criticism,
and painstaking efforts made it possible to present the work contained in this thesis in its
present form.
My sincere thanks goes to Dr. D. Kashyap, Head of the Civil Engineering Department, for
extending the computational and other facilities of the department.
I would like to take this opportunity to acknowledge the members of my examining

committee viz. Dr. I. V. Singh (Associate Professor, Department of Mechanical and
Industrial Engineering), Dr. P.K. Gupta (Associate Professor, Department of Civil
Engineering) and
Dr. A. Upadhaya (Professor & Chairman SRC, Department of Civil Engineering) for their
careful examinations of the work and their invaluable comments and insights, which made
a deep impact on my research.
Special thanks are due for my co-scholars in the Civil Engineering Department Ajay
Kumar, Tushar Kanti Dey, Sushil Kumar Singh, H.D. Chalak for extending all sorts of
helps without any hesitation and making my stay a very pleasant and memorable one
during my tenure at IIT, Roorkee.
I would like to thank the faculty and staff of the Civil Engineering Department, IIT
Roorkee for their immense help throughout the tenure of my research work.
I am also grateful to the faculty and staff of the Information Superhighway, IIT Roorkee
for providing me the required computational facilities.
iv
Last but certainly not least, my heartfelt thank go to my mother, brother and in-laws for
their encouragements and co-operations shown during this period. I would like to express
my warmest thanks to my husband S.M. Mohamed Hussain and my little daughter M.
Shiza Hussain, who endured all the sufferings silently and looked forward to this day.
Finally, I am thankful to the authorities of Indian Institute of Technology, Roorkee for

providing me all necessary assistance in the form of research and guidance.
Indian Institute of Technology (Gulshan Taj M.N.A)

Roorkee, India.
v
CONTENTS
Page No.
Abstract i-iii
Acknowledgement iv-v
Contents vi-xv
List of Figures xvi-xxiii
List of Tables xxiv-xxx
Nomenclature xxxi-xxxvi
CHAPTER 1 INTRODUCTION 1-10
1.1 Overview 1
1.2 Objectives and scope of the present research 9
1.3 Organization of the thesis 10
CHAPTER 2 REVIEW OF LITERATURE 11-80
2.1 Introduction 11
2.2 Brief review of various shear deformation theories of

plates/shells 12
2.3 Static, dynamic and buckling responses of FGM plates/shells 21
2.4 Static, dynamic and buckling responses of skew plates/shells 38
2.5 Static, dynamic and buckling responses of sandwich

plates/shells 45
2.6 Geometrically non-linear responses of laminated and

functionally graded plates/shells 58
2.7 Post buckling response of laminated and functionally graded

plates/shells 70
2.8 Appraisal from previous research works 77
2.9 Summary 79
vi
CHAPTER 3 MATHEMATICAL IDEALIZATION OF 84-132
FGM AND FE FORMULATION
3.1 Introduction 84
3.2 Homogenization of FGMs 85
3.2.1 Self consistent scheme 85
3.2.2 Mori-Tanaka approach 86
3.2.3 Composite sphere assemblage model 86
3.2.4 Composite cylindrical assemblage model 87
3.2.5 The simplified strength of materials method 87
3.2.6 The method of cells 87
3.2.7 Micromechanical models 87
3.3 Idealization of FGMs 87
3.3.1 The Exponential law 88
3.3.2 The Power law 88
3.4 Geometrical Configuration 88
3.4.1 Single layer FGM plate/shell 88
3.4.2 Sandwich FGM plate/shell 92
3.4.2.1 Sandwich FGM plate/shell with FGM skin 93
3.4.2.2 Sandwich FGM plate/shell with FGM core 95
3.5 Four-parameter power law distribution 97
3.6 Skew boundary transformation 98
3.7 Thermal analysis 99
3.8 Assumptions 100
3.9 Material modeling 100
3.9.1 Constitutive relationship 100
3.9.2 Element description 101
3.10 Displacement field 103
vii
3.10.1 Displacement field having constant variation of
103
transverse displacement
3.10.1.1 Strain-displacement relationship for plate 105
3.10.1.2 Strain-displacement relationship for shell 105
3.10.2 Displacement field having quadratic variation of

107
transverse displacement
3.10.2.1 Strain-displacement relationship for plate 108
3.10.2.2 Strain-displacement relationship for shell 110
3.10.3 Governing equation for static analysis 111
3.11 Governing equation for dynamic analysis 113
3.11.1 Free vibration analysis 113
3.11.2 Forced vibration analysis 114
3.12 Governing equation for buckling analysis 115
3.13 Geometrically non linear analysis 116
3.13.1 Green-Lagrange strain displacement relation 116
3.13.2 Solution to non-linear equilibrium equations 120
3.14 Formulation for post buckling analysis 123
3.14.1 Green-Lagrange strain-displacement relations 123
3.14.2 Strain energy of plate 127
3.14.3 Derivation of the governing equation 128
3.15 Boundary conditions 129
3.15.1 Boundary conditions considering constant transverse

129
displacement field
3.15.2 Boundary conditions considering quadratic variation of
130
transverse displacement field
3.16 Computer implementation 131
3.16.1 Introduction 131
3.16.2 Application domain 131
3.17 Summary 132
viii
CHAPTER 4 RESULTS AND DISCUSSIONS 133-290
4.1 Introduction 133
4.2 Static analysis of FGM skew plate/shell subjected to

mechanical and thermal loading 135
4.2.1 Convergence and comparison of results of Al/ZrO2 and

Al/SiC plates under uniformly distributed and sinusoidal
loading conditions 136
4.2.2 Convergence and comparison of results for Al/ZrO2

cylindrical shell subjected to mechanical load 139
4.2.3 Non-dimensional central deflection and normal stresses

for FGM plate under thermal loading 140
4.2.4 Non-dimensional central deflection and in-plane normal

stress for FGM skew plate subjected to mechanical and
thermal loading 141
4.2.5 Non-dimensional central deflection and normal stresses

for FGM skew shell under mechanical and thermal
loading 144
4.2.5.1 Non-dimensional central deflection and in-plane

normal stress for FGM skew cylindrical shell
under mechanical loading 144
4.2.5.2 Non-dimensional central deflection of Al/Al2O3

spherical, hypar and hyperbolic paraboloid skew
shell subjected to mechanical loading 145
4.2.5.3 Influence of R/h ratio on non-dimensional

deflection of Al/Al2O3 cylindrical skew shell
under mechanical loading 146
4.2.5.4 Effect of thermal loading on non-dimensional

deflection of Al/Al2O3 skew cylindrical shell 148
4.3 Free vibration of FGM skew plate/shell 149
4.3.1 Convergence and comparison of results of free vibration

frequency for FGM skew plate 149
4.3.1.1 Comparison of frequency values of SuS3O4/Si3N4

plate 149
4.3.1.2Non-dimensional frequency values of

SUS3O4/Si3N4 skew plate 150
4.3.2 Thermal vibration Frequency for FGM skew plate having 153
ix
different temperature dependent properties
4.3.2.1 Validation of non-dimensional free vibration

frequency values of SUS3O4/Si3N4 plate 153
4.3.2.2 Non-dimensional free vibration frequency

values of SuS3O4/Si3N4 skew plate 155
4.3.2.3 Influence of thermal field on non-dimensional

frequency values of SUS3O4/Si3N4 skew plate 157
4.3.2.4 Influence of aspect ratio on non-dimensional

free vibration frequency values of
Si3N4/SUS3O4 skew plate 159
4.3.2.5 Influence of thickness ratio on non-dimensional

frequency values of Si3N4/SUS3O4 skew plate 160
4.3.3 Free vibration frequency and forced vibration response

of FGM skew shell 161
4.3.3.1 Validation of free vibration frequency of

simply supported Si3N4/SUS3O4 shell 161
4.3.3.2 Validation of forced vibration response of

Al/ZrO2 plate 163
4.3.3.3 Influence of thickness ratio on free vibration

response of Al/SiC cylindrical shell 163
4.3.3.4 Influence of thickness ratio on free vibration

response of Al/SiC spherical shell 165
4.3.3.5 Influence of curvature ratio (R/a) on free

vibration frequency of Al/SiC cylindrical and
spherical shell 166
4.3.3.6 Free vibration frequency of simply supported
and clamped Al/SiC hypar shell 167
4.3.3.7 Forced vibration response of Si3N4/SUS3O4
167
skew shell
4.4 Buckling analysis of FGM skew plate subjected to in-plane
mechanical loading and thermal fields 170
4.4.1 Validation of the results of buckling analysis of

Al/Al2O3 skew plate subjected to uni-axial load 171
4.4.2 Validation of the results of buckling analysis of

isotropic and Al/Al2O3 FGM skew plate subjected to
thermal loading 173
x
4.4.3 Buckling analysis for FGM skew plate subjected to
uni-axial and bi-axial in plane loads 174
4.4.4 Influence of aspect ratio (a/b) on critical buckling

temperature of simply supported Al/Al2O3 skew plate
subjected to uniform temperature rise 176
4.4.5 Critical buckling temperature of simply supported

Si3N4/ SUS3O4 skew plate having thermal dependent
properties and subjected to uniform temperature rise 176
4.5 Static, free vibration and buckling analyses of FGM plate

described by four-parameter dependent power law distribution 177
4.5.1 Static analysis of Al/ZrO2 plate subjected to mechanical

loading considering different FGM profiles 178
4.5.2 Free vibration frequency of Al/ZrO2 plate by

considering different FGM profiles 185
4.5.2.1 Influence of boundary conditions and thickness

on free vibration analysis of four-parameter
Al/ZrO2 plate 185
4.5.2.2 Influence of material gradient parameters on

free vibration analysis of four-parameter
Al/ZrO2 plate 189
4.5.3 Buckling of four-parameter FGM plate subjected to

mechanical and thermal loading 193
4.5.3.1 Buckling of four-parameter Al/Al2O3 plate

subjected to mechanical loading 193
4.5.3.2 Buckling load for four-parameter

Si3N4/SUS3O4 plate subjected to thermal
loading 196
4.6 Static analysis of FGM sandwich skew plates and shells 197
4.6.1 Static analysis of FGM sandwich skew plates subjected

to mechanical loading 197
4.6.1.1 Convergence and validation study for FGM

sandwich plate subjected to mechanical loading 198
4.6.1.2 Influence of skew angle () and volume

fraction index (n) on displacement parameter
for Type A FGSSP 201
4.6.1.3 Influence of skew angle () and volume

204
fraction index (n) on displacement parameter
xi
for Type B FGSSP
4.6.1.4 Influence of boundary conditions on

displacement parameter for Type A and Type B
FGSSP 205
4.6.1.5 Through-the-thickness variation of in-plane

axial and shear stresses for Type A FGSSP 206
4.6.1.6 Through-the-thickness variation of in-plane axial

and shear stresses for Type B FGSSP 208
4.6.2 Static analysis of FGM sandwich skew shells subjected

to mechanical loading 212
4.6.2.1 Validation of results of static analysis of FGM

sandwich skew shells subjected to mechanical
loading 213
4.6.2.2 Influence of side-thickness ratio (a/h) on non-

dimensional central deflection of Type A and
Type B FGSSS 213
4.6.2.3 Influence of radius-side ratio (R/a) on non-

dimensional central deflection of Type A and
Type B FGSSS 216
4.6.2.4 Variation of non-dimensional in-plane normal

and shear stresses for Type A and Type B
Al/ZrO2 FGSSS 217
4.7 Free vibration analysis of FGM sandwich skew plates and

shells 221
4.7.1 Validation of free vibration results of FGSSP 221
4.7.1.1 Influence of boundary conditions on frequency

response of FGSP having graded and
homogeneous core 222
4.7.1.2 Influence of thickness-side ratio on frequency

parameter of FGSSP 226
4.7.2 Free vibration frequency of FGSSS by considering

different boundary conditions, thickness schemes and
skew angles 229
4.7.2.1 Influence of boundary conditions on frequency

parameter of FGSSS 230
4.7.2.2 Influence of thickness-side ratio on frequency

values of FGSSS 234
xii
4.8 Geometrically non-linear (GNL) analysis of FGSP and FGSS 237
4.8.1 GNL of FGSP by considering different thickness

schemes 237
4.8.1.1 Validation of results for geometrically non-

linear analysis of isotropic and FGM plates
4.8.1.2 Non-linear load-displacement response of FGSP

having graded core subjected to mechanical
loading 239
4.8.1.3 Influence of thickness ratio (a/h) on non-linear

load-displacement response of FGSP having
graded core subjected to mechanical loading 241
4.8.1.4 Non-linear load-displacement response of FGSP

having homogeneous core subjected to
mechanical loading 242
4.8.1.5 Influence of thickness ratio (a/h) on non-linear

load-displacement response of FGSP having
homogeneous core subjected to mechanical
loading 245
4.8.2 Geometrically non-linear analysis of FGM cylindrical,

spherical, hyperbolic paraboloid, and hypar shells
4.8.2.1 Validation of results for Al/ZrO2 FGM

cylindrical shell subjected to mechanical
loading 246
4.8.2.2 Geometrically non-linear analysis of FGM

cylindrical shells subjected to mechanical
loading 247

spherical shells subjected to mechanical
loading 250

hyperbolic paraboloid shells subjected to
4.8.2.5 Geometrically non-linear analysis of FGM hypar

shells subjected to mechanical loading 254
4.8.3 Geometrically non-linear analysis of cylindrical,

spherical, hypar and hyperbolic paraboloid FGM
sandwich shells subjected to mechanical loading 256
xiii
4.8.3.1 Influence of thickness scheme on non-linear
bending response of FGM cylindrical
sandwich shells subjected to mechanical
loading 256
4.8.3.2 Influence of curvature (1/Rx) on non-linear

bending of cylindrical FGSP subjected to
4.8.3.3 Variation of in-plane normal stresses (𝛔xx) in

cylindrical FGSS subjected to non-linear
bending 263
4.8.3.4 Geometrically non-linear response of FGM

spherical sandwich shells subjected to
4.8.3.5 Geometrically non-linear response of FGM

hyperbolic paraboloid sandwich shells
4.8.3.6 Geometrically non-linear response of hypar

sandwich shells subjected to mechanical
loading 279
4.9 Post buckling analysis of FGSP 283
4.9.1 Validation of post buckling analysis of FGSP 284
4.9.2 Non-dimensional critical buckling load of FGSP for

different amplitude ratios 284
4.9.3 Influence of volume fraction index (n) on post buckling

strengths of FGSP 287
4.9.4 Influence of aspect ratio (b/a) on post buckling

4.9.5 Influence of boundary conditions on post buckling

4.10 Summary 290
CHAPTER 5 CLOSURE 291-297

5.1 Summary and Conclusions 291
5.1.1 Summary 291
5.1.2 Conclusions 293
xiv
5.1.3 Major Contributions 296
5.2 Scope of future research 297
REFERENCES 298-332
APPENDIX 333-345
A.1 Transformation matrix for constant variation of transverse
displacement field 333
A.2 Transformation matrix for quadratic variation of transverse
displacement field 333
B.1 [H]p1 matrix for FGM plate considering constant variation of
transverse displacement field 334
B.2 [H] p2 matrix for FGM plate considering quadratic variation of
B.3 [H]G matrix for FGM plate considering constant variation of
B.4 [H]G matrix for FGM plate considering quadratic variation of
C.1 Non-linear strain terms 336
D.1 Different terms of matrix [A] 341
LIST OF PUBLICATIONS 346-348
xv
LIST OF FIGURES
Figure Figure Caption Page
No. No.
1.1 Chart showing various potential areas of application of FGMs 3
3.1 A Schematic representing single layer FGM plate 89
3.2 A Schematic representing single layer FGM cylindrical shell 89
3.3 Through-the-thickness variation of volume fraction of ceramic (Vc) for

different n values 90
3.4 Through-the-thickness variation of Young’s modulus (E) for different n

values 91
3.5 Through-the-thickness variation of thermal expansion (𝛂) for different n

values 92
3.6 Geometry of FGM sandwich having graded skin layers and isotropic
core 93

different types of sandwich plates having FGM skin 94
3.8 Geometry of FGM sandwich having graded core and isotropic face
sheets 95
Through-the-thickness variation of volume fraction of ceramic (Vc) for
3.9
different types of sandwich plates having FGM core 96

single layer FGM plate/shell according to four-parameter power law
distribution 97
3.11 Geometry of single layer FGM skew plate/shell 99
3.12 Isoparametric nine node Lagrangian element in natural co-ordinate

system 102
3.13 Skew plate co-ordinate system 130
4.1 Influence of boundary conditions on non-dimensional central deflection 146
xvi
of simply supported Al/Al2O3 square shell for different skew angles
subjected to mechanical loading (h=0.01m)
4.2 Influence of radius-thickness ratio on non-dimensional central

deflection of simply supported Al/Al2O3 square cylindrical shell for
different skew angles subjected to mechanical loading (h=0.01m) 147
4.3 Influence of load parameter on non-dimensional central deflection of

simply supported square Al/Al2O3 cylindrical shell for different volume
fraction index subjected to mechanical loading (h=0.01m) 147
4.4 Influence of boundary conditions on frequency parameter of

SuS3O4/Si3N4 square plate for different skew angles (h=0.1m and n=1.0) 150
4.5 Influence of boundary condition on frequency parameter of square

SUS3O4/Si3N4 plate for different skew angles (h=0.2m and n=1.0) 151
4.6 Influence of boundary conditions on frequency parameter of

SUS3O4/Si3N4 plate for different skew angles (b/a=2.0, h=0.1m and
n=1.0) 152
4.7 Influence of volume fraction index on frequency parameter of

SUS3O4/Si3N4 plate for different skew angles (b/a=2.0, h=0.1m and
n=0) 153
4.8 Influence of aspect ratio (a/b) on frequency parameter of simply

supported Si3N4/SUS3O4 plate for different skew angles (h=0.1m) 159
4.9 Influence of aspect ratio (a/b) on frequency parameter of clamped

Si3N4/SUS3O4 plate for different skew angles (h=0.1m) 160
4.10 Influence of thickness ratio on frequency parameter of square clamped

ZrO2/Ti-6Al-4V plate for different skew angles 161
4.11 Non-dimensional displacement of simply supported Al/ZrO2 square

plate for different values of volume fraction index 163
4.12 Influence of R/a ratio on non-dimensional frequency of Al/SiC

cylindrical skew shell (n =1.0, a/h=10) 166
4.13 Influence of R/a ratio on non-dimensional frequency of Al/SiC spherical

skew shell (n =1.0, a/h=10) 166
xvii
4.14 Influence of skew angle on the dynamic response of Si3N4/SUS3O4
cylindrical skew shell (n =1.0, a/h=10) 168
4.15 Influence of aspect ratio (b/a) on transient response of Si3N4/SUS3O4

cylindrical skew shell (=15°, n=1.0, a/h=10) 169
4.16 Influence of shell geometry on the dynamic response of Si3N4/SUS3O4

skew shell (=15°, n =1.0, a/h=10) 169
4.17 Influence of volume fraction index (n) on the dynamic response of

Si3N4/SUS3O4 cylindrical skew shell (=15°, a/h=10) 170
4.18 Non-dimensional in-plane normal stresses of Si3N4/SUS3O4 cylindrical

skew shell (n =1.0, a/h =100) 170
4.19 Influence of aspect ratio (a/b) on non-dimensional buckling temperature

of simply supported Al/Al203 plate under uniform temperature rise
(a/h=10 and n=1) 176
Non-dimensional deflections of Al/ZrO2 plate having different boundary
4.20
conditions based on four-parameter power law function (a/h=10) 178
4.21 Influence of thickness ratio (a/h) on non-dimensional deflection of

Al/ZrO2 plate for different volume fraction index based on four-
parameter power law function 179
4.22 Influence of aspect ratio (a/b) on non-dimensional deflection of Al/ZrO2

plate for different volume fraction index based on four-parameter power
law function (a/h=10) 180
4.23 Through-the-thickness variation of in-plane normal stress (σxx) of

parameter power law function (a/h=10) 181
4.24 Through-the-thickness variation of in-plane normal stress (σyy) of

4.25 Through-the-thickness variation of in-plane shear stress (𝛕xy) of

xviii
4.26 Through-the-thickness variation of transverse shear stress (𝛕yz) of
4.27 First six natural frequencies of four-parameter Al/ZrO2 FGM plate with
varying b1 and c1 values (0<a1<1.2) 190
varying a1 and c1 values (0<b1<1.2) 191
varying a1 and b1 values (1<c1<11) 192
4.30 FGM plate subjected to various in-plane loading conditions 194
4.31 Variation of critical buckling load (Ncr) of four-parameter Al/Al2O3

plate for different volume fraction index (n) under different conditions
of in-plane edge loads 195
4.32 Influence of thickness-side (h/a) ratio on critical buckling load (Ncr) of

four-parameter Al/Al2O3 plate subjected to bi-axial compression 195
Influence of aspect ratio (a/b) on critical buckling load (Ncr) of four-
4.33
parameter Al/Al2O3 plate subjected to bi-axial compression 196
4.34 Through-the-thickness variation of in-plane normal stress (xx) for

different thickness schemes of Type A FGSSP 206
4.35 Through-the-thickness variation of in-plane normal stress (yy) for

4.36 Through-the-thickness variation of in-plane shear stress (𝛕xy) for


different thickness schemes of Type B FGSSP 210


4.40 Influence of thickness ratio (a/h) on non dimensional central deflection 214
xix
of Al/ZrO2 Type A FGSSS
4.41 Influence of thickness ratio (a/h) on non dimensional central deflection

of Al/ZrO2 Type B FGSSS 215
Influence of curvature-side ratio (R/a) on non-dimensional central
4.42
deflection of Al/ZrO2 Type A FGSSS 216
4.43 Influence of curvature-side ratio (R/a) on non-dimensional central

deflection of Al/ZrO2 Type B FGSSS 217

different thickness schemes of Type A FGSSS 218


4.47 Through-the-thickness variation of in-plane normal (xx and yy) and

shear stresses (𝛕xy) for different thickness schemes of Type B FGSSS
(n=1) 220
4.48 Validation of non-linear load-displacement response of isotropic plates 238
4.49 Validation of non-linear load-displacement response of FGM plates 239
4.50 Non-linear load-displacement response of FGSP having graded core by

considering various thickness schemes and volume fraction indexes (n)
subjected to loading magnitude of 1-5 N/mm2 239
4.51 Non-linear load-displacement response of FGSP having graded core by

subjected to load magnitude of 100-1000 N/mm2 240
4.52 Influence of thickness ratio (a/h) ratio on non-linear bending response

of FGSP having graded layer as core 242
4.53 Non-linear load-displacement response of FGSP having homogeneous

core by considering various thickness schemes and volume fraction
indexes (n) subjected to load magnitude of 1-5 N/mm2 243
4.54 Non-linear load-displacement response of FGSP having graded core by 244
xx
subjected to load magnitude of 100-1000 N/mm2
4.55 Influence of thickness ratio (a/h) on non-linear bending response of

FGSP having graded layer as core 245
4.56 Non-linear load-deflection graph for GNL of FGM cylindrical shells

Non-linear load-displacement curves of FGM cylindrical shells for
4.57
different magnitude of loading and boundary conditions 248
4.58 Influence of radius on non-linear deflection response of clamped FGM

cylindrical shells for different ranges of volume fraction index (n) 249
4.59 Through-the-thickness variation of in-plane axial stress (𝛔xx) of FGM

cylindrical shells for different boundary conditions and thickness values
under non-linear bending 249
4.60 Non-linear load-displacement curves of FGM spherical shells for

different magnitude of loadings and boundary conditions 250

spherical shells for different ranges of volume fraction index (n) 250
4.62 Through-the-thickness variation of in-plane normal stresses (𝛔xx) of

FGM spherical shells for different boundary conditions and thickness
value under non-linear bending 251
4.63 Non-linear load-displacement curves of FGM hyperbolic paraboloid

shells for different magnitude of loading and boundary conditions 252

hyperbolic paraboloid shells for different values of volume fraction
index (n) 253

FGM hyperbolic paraboloid shells 253
4.66 Non-linear load-displacement curves of FGM hypar shells for different

magnitude of loading and different boundary conditions 254
4.67 Influence of radius on non-linear deflection response of clamped FGM 255
xxi
hypar shells for different ranges of volume fraction index (n)

FGM hypar shells 255
4.69 Non-linear load-deflection behaviour of FGSS (CSG) 257
4.70 Non-linear load-deflection behaviour of FGSS (CSH) 258
4.71 Non-linear load-deflection equilibrium path of FGSS (CSG) (CCCC

boundary) 259
4.72 Non-linear load-deflection equilibrium path of FGSS (CSH) (CCCC

boundary) 260
4.73 Influence of curvature on non-linear displacement for CSG (CCCC

boundary) 262
4.74 Through-the-thickness in-plane stress (𝛔xx) variation of CSG (CCCC

boundary) 263
4.75 Through-the-thickness in-plane stress (𝛔xx) variation of CSH (CCCC

boundary) 264
4.76 Non-linear load-deflection graph for FGSS (SPG) (CCCC boundary) 266
4.77 Non-linear load-deflection graph for FGSS (SPH) (CCCC boundary) 267
4.78 Non-linear load-deflection graph for FGSS (SPG) (SSSS boundary) 269
4.79 Non-linear load-deflection graph for FGSS (SPH) (SSSS boundary) 270
Influence of curvature on non-linear displacement for SPG sandwich
4.80
shells (CCCC boundary) 271
4.81 Through-the-thickness in-plane normal stress (𝛔xx) variation for SPG

sandwich shells (CCCC boundary) 272
4.82 Through-the-thickness in-plane normal stress (𝛔xx) variation for SPH

sandwich shells (CCCC boundary) 273
4.83 Non-linear load-deflection equilibrium path for FGSS (HYPG) (CCCC

boundary) 275
4.84 Non-linear load-deflection equilibrium path for FGSS (HYPH) (CCCC

boundary) 276
xxii
4.85 Through-the-thickness in-plane stress (𝛔xx) variation for FGSS (HYPG)
(CCCC boundary) 277
4.86 Through-the-thickness in-plane stress (𝛔xx) variation for FGSS (HYPH)

(CCCC boundary) 278
4.87 Non-linear load-deflection equilibrium path for FGSS (HYG) (CCCC

boundary) 280
4.88 Non-linear load-deflection equilibrium path for FGSS (HYH) (CCCC

boundary) 281
4.89 Influence of curvature on non-linear displacement for HYG (CCCC

boundary) 282
4.90 Through-the-thickness in-plane normal stress (𝛔xx) variation for HYG

(CCCC boundary) 283
4.91 Post buckling response of Al/ZrO2 plate under bi-axial compression

(a/h=40, a/b=1, SSSS boundary) 284
4.92 Influence of volume fraction index (n) on post buckling strength of

different FGSP-G for different amplitude ratios (w/h) 288
Influence of aspect ratio (b/a) on post buckling strength of different
4.93
FGSP-G for different amplitude ratios (w/h) 288
4.94 Influence of aspect ratio (b/a) on post buckling strength of different

FGSP-H for different amplitude ratios (w/h) 289
4.95 Influence of boundary conditions on post buckling strength of FGSP-G

for different amplitude ratios (w/h) 289
4.96 Influence of boundary conditions on post buckling strength of FGSP-H

for different amplitude ratios (w/h) 290
xxiii
LIST OF TABLES
Table Table Caption Page
No. No.
2.1 Summary of the important research works 81
4.1 Mechanical properties of different FGM constituents 135
4.2 Convergence and comparison of results of Al/Zr02 plate having
different thickness ratio (a/h) and boundary conditions 137
4.3 Convergence and comparison of results of Al/SiC square plate having
different thickness ratio (a/h) and boundary conditions 138
4.4 Convergence and comparison of results of simply supported Al/SiC
FGM plate having different aspect ratios (b/a) (a/h=5.0) 138
4.5 Non-dimensional normal stresses at the center of simply supported
Al/ZrO2 square plate for different skew angles subjected to uniformly
distributed load (h=0.1) 139
4.6 Convergence and comparison of results of Al/ZrO2 FGM square shell
for different boundary conditions and volume fraction index 139
4.7 Effect of thermal loading on non dimensional normal stresses of simply
supported Al/ZrO2 square plate for different volume fraction index and
thickness ratio (a/h) 140
4.8 Effect of thermal loading on non-dimensional central deflection of
simply supported Al/ZrO2 square plate for different volume fraction
index and thickness ratio (a/h) 141
4.9 Non-dimensional central deflection of simply supported Al/ZrO2 square
plate for different volume fraction index and skew angles subjected to
4.10 Non-dimensional normal stresses at the centre of simply supported
Al/ZrO2 square plate for different volume fraction index and skew
angles subjected to mechanical loading 142
4.11 Non-dimensional central deflection of simply supported Al/ZrO2 square
plate for different volume fraction index and skew angles subjected to
thermal field 143
4.12 Variation of non-dimensional normal stresses at the centre of simply 143
xxiv
supported Al/ZrO2 square plate for different volume fraction index and
skew angles subjected to thermal loading
4.13 Non-dimensional central deflection of square Al/ZrO2 cylindrical shell
for different skew angles and boundary conditions (h=0.01m) 144
4.14 Non-dimensional normal stresses of simply supported Al/ZrO2 square
cylindrical shell for different skew angles and volume fraction index
(h=0.01m) 145
4.15 Non-dimensional central deflection of SUS3O4/Si3N4 square cylindrical
shell for different volume fraction index and skew angles subjected to
thermal field (h=0.01m) 148
4.16 Non-dimensional frequency parameter of simply supported
SuS3O4/Si3N4 square skew plate (h=0.1m) 150
4.17 Temperature dependent properties for different FGM constituents at
T=300K (Reddy and Chin 1998) 154
4.18 Validation table showing frequency values of Si3N4/ SUS3O4 plate for
different values of volume fraction index 155
4.19 Non-dimensional frequency values for simply supported Si3N4/ SUS3O4
square plate for different values of skew angles and volume fraction
index (a/h=10) 156
4.20 Non-dimensional frequency values for clamped Si3N4/ SUS3O4 square
plate for different values of skew angles and volume fraction index
(a/h=10) 156
4.21 Non-dimensional frequency values for simply supported-clamped
Si3N4/ SUS3O4 square plate for different values of skew angles and
volume fraction index (a/h=10) 157
4.22 Influence of thermal field on non-dimensional frequency values of
Si3N4/ SUS3O4 square plate for different values of skew angles and
boundary conditions subjected to thermal field (h=0.1m) 157
4.23 Influence of thermal field on non-dimensional frequency values of
simply supported ZrO2/ Ti-6Al-4V square plate for different values of
skew angles (h=0.1m) 158
4.24 Validation study for free vibration of simply supported Si3N4/SUS3O4
cylindrical shell (a/h=10) 162
4.25 Non-dimensional frequency values of clamped Al/SiC square 164
xxv
cylindrical shell (R/a=5.0)
4.26 Non-dimensional frequency values of simply supported Al/SiC square
cylindrical shell (R/a=5.0) 164
4.27 Non-dimensional frequencies of clamped Al/SiC square spherical shell 165
4.28 Non-dimensional frequencies of simply supported Al/SiC square
spherical shell 165
4.29 Non-dimensional frequency values of square Al/SiC hypar shell
(a/h=10.0) 167
4.30 Convergence and comparison study of critical buckling load for simply
supported Al/Al2O3 square skew plate subjected to uni-axial in-plane
load ( h=0.01m) 171
4.31 Convergence and comparison of results of critical buckling load for
simply supported Al/Al2O3 square skew plate subjected to uni-axial in-
plane load ( h=0.01m) 172
4.32 Non-dimensional critical buckling load of square isotropic plate
subjected to uni-axial and bi-axial in-plane loads (h=0.01m) 173
4.33 Non-dimensional critical buckling temperature of Al/Al2O3 square plate
under uniform temperature rise 174
4.34 Non-dimensional critical buckling load for simply supported Al/ZrO2
square skew plate subjected to uni-axial and bi-axial in-plane loads
( h=0.1 m) 174
4.35 Non-dimensional critical buckling load of clamped Al/Al2O3 plate
under in-plane loading for different skew angles (Ψ) 175
4.36 Critical buckling temperature (ΔTcr) of simply supported Si3N4/ SUS3O4
skew plate for various skew angles (a/h =10 and n = 1.0) 177
4.37 Different profiles and values of material gradient parameters 177
4.38 Natural frequencies of simply supported four-parameter FGM plate for
first six modes (h=0.1m) 186
4.39 Natural frequencies of simply supported four-parameter FGM plate for
first six modes (h=0.01m) 186
4.40 Natural frequencies of clamped four-parameter FGM plate for the first
six modes (h=0.1m) 187
4.41 Natural frequencies of clamped four-parameter FGM plate for first six
modes (h=0.01m) 187
xxvi
4.42 Natural frequencies of simply supported-clamped four-parameter FGM
plate for the first six modes (h=0.1m) 188
4.43 Natural frequencies of simply supported-clamped four-parameter FGM
plate for the first six modes (h=0.01m) 188
4.44 Critical buckling temperature (∆Tcr) of four-parameter Si3N4/SUS304
plate under uniform temperature rise 197
4.45 Critical buckling temperature (∆Tcr) of four-parameter Si3N4/SUS3O4
plate under non-linear temperature field (°C) 197
4.46 Non-dimensional displacements for Type A Al/ZrO2 FGSP for different
thickness schemes and volume fraction indexes (n) 199
4.47 Validation table for Type B Al/Al2O3 FGM sandwich plate having
different thickness schemes and various volume fraction index (n) 200
4.48 Non-dimensional central deflection of Type A FGSSP for various skew
angle () and volume fraction index (n) (a/h=4) 201
4.51 Non-dimensional transverse displacement of Type B FGSSP for
different skew angle () and volume fraction index (n) 204
4.52 Influence of aspect ratio (a/b) on non-dimensional deflection (w) of
Type A and Type B FGSSP for various skew angle ( ) and volume
fraction index (n) 205
4.53 Non-dimensional values of in-plane normal and shear stresses for
different values of skew angles and thickness scheme for Type A
FGSSP 209
4.54 Non-dimensional in-plane normal and shear stresses for different values
of skew angles and thickness scheme for Type B FGSSP 212
4.55 Deflection of Al/ZrO2 shell for various radius-thickness ratios (R/h) and
volume fraction indexes (n) 213
4.56 Validation study for free vibration of FGSP for different thickness and
volume fraction index values 222
4.57 Influence of boundary conditions on frequency parameter of FGSSP 222
xxvii
having graded core (=15°)
4.58 Influence of boundary conditions on frequency parameter of FGSSP
having graded core (=30°) 223
having homogeneous core (=15°) 224
4.65 Influence of thickness-side ratio (h/b) on non-dimensional frequency
values of FGSSP having graded core for different volume fraction
index and skew angle values (Type 1) 226
values of FGSSP having homogeneous core for different volume
fraction index and skew angle values (Type 1) 228
4.71 Influence of thickness-side ratio (h/b) on non-dimensional frequency 229
xxviii
fraction index and skew angle values (Type 3)
4.73 Influence of boundary conditions on frequency values of FGSSS having
graded core (=15°) 230
homogeneous core (=15°) 232
values of FGSSS having graded core for different volume fraction
xxix
values of FGSSS having homogeneous core for different volume
4.89 Load ratio (Nb/Ncr) of FGSP for different thickness schemes and
amplitude ratios (a/h=100) 285
4.90 Load ratio (Nb/Ncr) of FGSP for different thickness schemes and
amplitude ratios (a/h=5) 286
xxx
NOMENCLATURE
a Dimension of the plate along x - axis
a1,b1,c1 Material gradient parameters
b Dimension of the plate/shell along y-axis
B Bulk modulus
[B] Strain-displacement matrix
 B L Linear strain-displacement matrix
 B NL Non-linear strain-displacement matrix
[C] Damping matrix

C1 Tracer in Sander’s shell theory
dS Undeformed surface
[D] Rigidity matrix
Dc Rigidity value of ceramic
E Young’s modulus
E0 Reference value of Young’s modulus of metal
Ec Young’s modulus of ceramic
Em Young’s modulus of metal
Emax Maximum stiffness
{f} Acceleration vector
F(t) Dynamic load shape function
h Total thickness of plate/shell
ho Bottom surface of FGM/homogeneous skin layer
h1 Bottom surface of homogenous/FGM core layer
h2 Top surface of homogenous/FGM core layer
h3 Top surface of FGM/homogeneous skin layer
k Heat conductivity co-efficient
[k] Element stiffness matrix
[kg] Element geometric stiffness matrix
K Kelvin
[Kg] Global geometric stiffness matrix
[K0] Initial stiffness matrix
[KL] Linear stiffness matrix
xxxi
[KNL] Non-linear stiffness matrix
[K𝜎] Initial stress matrix
 KT  Tangent stiffness matrix
[K] Global Stiffness matrix

[K’] Effective stiffness matrix
[m] Element mass matrix
[M] Global mass matrix
n Volume fraction index
Nb Primary post buckling load
Ncr Critical post buckling load
Ni Shape functions
[Nw]T Shape function matrix
{p} Load vector
{P} Surface traction
Pc, Pm Material property of ceramic and metal, respectively
P0, P-1, P1, P2, P3 Temperature Co-efficients
q Intensity of applied load
q0 Maximum dynamic amplitude
Qij Stiffness co-efficients
R Equivalent nodal load vector due to body forces and
tractions
Rxy Twist radii of curvature
[R’] Effective load
S i  Stress matrix
 
[T] Transformation matrix

T c , Tm Temperature of ceramic and metal, respectively
u, v, w Displacement components along x, y and z axes,
respectively
{u}i Displacement at node i
u0, v0, w0 Mid-plane displacement components along x, y and z axes,
respectively
Us Strain energy of the plate
xxxii
 .  Velocity
U 
 
 ..  Acceleration
U 
 
Vc Volume fraction of ceramic

Vm Volume fraction of metal
Vc(1), Vc(2), Vc(3) Volume fraction of bottom, core and top layers,
respectively
W Work done by the in-plane loads
We Potential of the externally applied load
x’, y’ Local axes system of skew plate along x and y directions,
respectively
z Thickness co-ordinate of plate/shell
α and β Damping constants
αm Thermal-expansion of metal
αc Thermal-expansion of ceramic
γxz, γyz Transverse shear strains along xz and yz planes,
respectively
γ1 Convergence tolerance
 Critical buckling load
1 Exponential parameter
σxx In-plane axial stress in x-direction
σyy In-plane axial stress in y-direction
σzz Transverse normal stress in z-direction
τxy In-plane shear stress in xy-plane
τxz In-plane shear stress in xz-plane
τyz In-plane shear stress in yz-plane
∆T Temperature rise
,  Natural co-ordinate system
ξx, ξy, x and y Higher order terms of Taylor’s series expansion
 Skew angle
,C Polynomial constants in heat conduction equation
θx, θy Bending rotations about y and x axes, respectively
ϵzz Strain component in z-direction
xxxiii
ρm Density of metal
ρc Density of ceramic
γm Poisson’s ratio of metal
ρ0 Reference value of density of metal
 Non-dimensional frequency value
ΔTcr Critical buckling temperature
∆t Time step in dynamic analysis
{ε} Total strain vector
{ε}m Mechanical strain vector
{ε}T Thermal strain vector
 L Linear strain vector
 NL Non-linear strain vector
{σ} Stress vector

{dε} Virtual Green’s strain vector
{du} Virtual displacement vector
{ 𝜀 }G Geometric strain vector
d  Variation of strain vector
du Variation of displacement vector
{θ} Displacement gradient

 i Incremental displacements at node i
Subscripts
c Ceramic
m metal
Abbreviations
CSG Cylindrical shell with graded core
CST Classical shell theory
CSH Cylindrical shell with homogeneous core
COV Co-efficient of variation
CPT Classical plate theory
DSC Discrete singular convolution
xxxiv
DRES Defense Research Establishment
DQEM Differential quadrature element method
FE Finite element formulation
FG Functionally graded
FGCP Functionally graded circular plates
FGP Functionally graded plate
FGSP Functionally graded sandwich plate
FGSSP Functionally graded skew sandwich plate
FGSS Functionally graded sandwich shell
FGSSS Functionally graded skew sandwich shell
FGM Functionally graded material
FGSP-H Functionally grades sandwich plate with homogeneous core
FGSP-G Functionally grades sandwich plate with graded core
FSDT First order shear deformation theory
GDQ General differential quadrature
GIQ Generalized integral quadrature
GNL Geometric non-linear
GPa Giga Pascal
HSDT Higher order shear deformation theory
HYPG Hyperbolic paraboloid with graded core
HYPH Hyperbolic paraboloid with homogeneous core
HYG Hypar shells with graded core
HYH Hypar shells with homogeneous core
IG Iso geometric
LS Least square
L4 Lagraningan element with 4 degrees of freedom
MTS Mori-Tanaka scheme
N-R Newton Raphson
NURBS Non uniform rational B-spline
PVD Principle of virtual displacements
QH Quadratic Heterosis
RMVT Reissner’s mixed variation theorem
RPT Refined plate theory
RVE Representative volume element
xxxv
SPG Spherical shell with graded core
SPH Spherical shell with homogeneous core
SSDT Sinusoidal shear deformation theory
TD Temperature-dependent
TID Temperature-independent
TSDT Third order shear deformation theory
VRM Voigt rule of mixture
1D One-dimensional
2D Two-dimensional
3D Three-dimensional
xxxvi
CHAPTER 1
INTRODUCTION
1.1 OVERVIEW
A general layout of composite structure consists of many constituent layers of the lamina
bonded together with at least two different kinds of materials at macroscopic scale.
Lamination is being performed to unite the superior aspects of the materials present in
constituent layers and bonding material, so as to achieve a more functional material that
satisfy the design requirements. In spite of the above merits, mismatch of mechanical/
thermal properties exists at layer interfaces often made them to sustain delamination and
debonding types of failure modes, especially under high operating environments. In
addition, the cracks may appear at the layer interfaces which generally propagate into
weaker direction or lamina and lead to stress concentrations at the layer interfaces under
external loading conditions.
The aforementioned drawbacks experienced by conventional composite laminates

paved the path to discover a new kind of materials, where these problems can be addressed
in an optimum/efficient manner. As a consequence, during a space plane project in 1984 at
National Aerospace Laboratory of Japan, a group of scientists discovered a class of
advanced composite materials named as functionally graded materials (FGM). These
advance materials with engineered gradients of composition, structure and/or specific
properties in the preferred direction/orientation are superior to corresponding
homogeneous material composed of identical constituents (Koizumi 1993). Though the
concept of FGMs and their fabrication technology appears to be an engineering invention,
the concept is not new. These kinds of materials are practically accessible in plenty of
forms such as bamboo tree, human skin and human bone. Although several numbers of
spatial and chemical configurations are possible in FGMs, in common, they involve a
blend of two distinct isotropic material phases. The resultant material is microscopically
heterogeneous and hence introduces the gradients by means of spatial variation of material
properties along the specified co-ordinates. Meanwhile, proper care has to be exercised
while tailoring these property gradients so as to achieve the benefit of two constituents.
1
In most of the research studies performed so far, FGM structure constituted by the
combination of metal and ceramic are accepted for their excellent outputs. Usually, large
concentrations of ceramic material are deposited at high temperature locations, while large
concentrations of metal are deposited at regions where mechanical strength is of great
concern. The region in between these large concentrations of ceramic and metal thus
results in smooth and continuous variation of desired properties
(mechanical/thermal/electrical) in the chosen direction.
The thermal conductivity of ceramics is very low compared with that of

corresponding metal part. As a result uneven stress distribution and expansion may occur
in a structure leading to fracture. In such situations, to overcome the brittle character of the
ceramic component, metal is combined with ceramics. This mixture of two distinct
isotropic materials is probable with two alternative options. The first choice is the
introduction of metal layers into ceramic structure, but it introduces interfaces and thus
results in disparity of material properties. Again this result into large concentration of
stresses, which is the undesirable effect often encountered in conventional composites. As
an alternative approach, a mixture of ceramic with the metal that exhibit a smooth and
continuously varying proportion of volume content seems to be better configuration at this
point. Thus the resulting structural component (i.e., FGM) may have gradual variation of
material properties without any interfaces and able to withstand ultra high temperature in
addition to high fracture toughness.
A general FGM structure usually refers to particulate composites where the volume
fraction of constituents varies in one or more directions (Birman and Byrd 2007). FGM
may also be formed incorporating fiber-reinforced layers with the volume fraction of
constituent phases as coordinate dependent instead of being constant and thus producing
the optimal set of structural response (Birman 1995; Birman 1997). While particulate type
FGMs may be isotropic at local domain, they may also exhibit heterogeneous properties
due to the spatial distribution of volume fraction of the constituents. Besides the particulate
type of FGMs, skeletal/orthotropic microstructure may be also manufactured using plasma
spray and electron beam physical vapor deposition techniques. In some situations, FGM
may include ceramics and two different metallic phases where the gradual variation of
material properties is achieved in the thickness direction according to power law
2
distribution. Such a configuration has proven to be very effective in reducing thermal
stresses when compared to conventional two-phase materials (Nemat Alla 2003).
With the potential to reduce the in-plane and transverse stresses, to improve
residual stress distribution, to enhance thermal properties, to give high fracture toughness,
and to reduce stress intensity factors, FGM components registered wide applications in
many engineering and other allied fields. Supersonic flight jets, rocket components, heat
exchanger tubes, biomedical implants, flywheels and plasma facings are some of the
notable fields that dominate in the list of applications.
Nuclear Projects Space Projects
(Fuel pellets, Plasma wall of (Rocket components, Space

fusion reactor) plane frames)
Communication field
Medical field
(Optical fibers, Lenses, FGM
Semiconductors generators, (Artificial bones, Skins,
Solar cells) Sensors) Dentistry)
Energy Sector Miscellaneous

(Thermoelectric generators,
(Building materials, Sport
Solar cells, Sensors)
goods, Window glasses)
Figure 1.1 Chart showing various potential areas of application of FGMs (Miyamoto 1999)
With superior thermal properties, FGM materials are proven to be capable of resisting a
temperature of 2000K with a temperature gradient of 1000K across a section of 10mm
thickness (Jha et al. 2013b). Over a course of time, these materials have expanded their
applications into chemical plants, solar energy generators, heat exchangers, nuclear and
chemical reactors, and high efficiency combustion systems. To improve thermal, oxidation
and corrosion properties they further extend their application as coatings in thermal
barriers systems. In thermo-electric field, the concept of functional grading has been
successfully implemented in sensors and thermo generators having metal-semiconductor
3
transition with improved efficiency. A brief chart showing the applications of FGMs in
diverse areas has been demonstrated in Figure 1.1.
The primary step in any analysis involves the assumption of suitable kinematics
field to accurately predict the response of the system under various operating
environments. In this context, to obtain the realistic response of the structure, many
displacement based theories are developed and proposed in the literatures. Classical plate
theory (CPT) proposed by Kirchhoff (1850) is the first theory which was implemented by
many researchers for the analysis of thin plate/shell structures. But CPT neglects the effect
of shear deformation and further assumes that a normal to the mid-plane remains straight
and normal to the mid-surface after deformation. As a consequence, the CPT usually
underestimates the deflection parameter and overestimates the natural frequencies and
buckling loads especially for thick plates (Reddy 2004). Also, this theory may be highly
unsuitable for structures made of FGM due to the phenomenon of continuous and gradual
volume distribution of two materials observed from point to point in the specific direction.
The major limitation of this theory has been realized during the analysis of thick plates,
where the contribution of shear deformation cannot be neglected. To propose an alternative
solution, a theory which considers the variation of shear deformation in linear sense is
introduced by Reissner-Mindlin (Reissner 1945b; Mindlin 1951) as first order shear
deformation theory (FSDT). But, the linear assumption of shear deformation variation
leads to the use of shear correction factor in order to account for the realistic parabolic
variation of transverse shear strain. As cited in most of the literatures, these factors are
very sensitive to the geometric properties, loading and boundary conditions. To address the
issues related with CPT and FSDT, many higher order shear deformation theories were
proposed (Reddy 1984b; Lo et al., 1967) to accomplish the realistic parabolic variation of
transverse stresses through the thickness. Generally HSDT involve the higher order term in
the Taylor’s expansion of the displacement component along the thickness direction.
Recently, different forms of HSDT models were proposed in the literature incorporating
sinusoidal, cubic, hyperbolic and exponential variations in the in-plane fields with constant
variation of transverse displacement (Thai and Choi 2013b) through the thickness, while
many others adopted the quadratic and cubic variation of transverse component in addition
to cubic variation of in-plane part for the analysis of FGM structures (Jha et al. 2012b; Jha
et al. 2013c; Talha and Singh 2010).
4
Skew plates belongs to the quadrilateral plate family are often identified as
parallelogram plate, swept plate, rhombic plate and oblique plate and serve as major load
bearing components in many structures. FGM skew plates have wide applications in areas,
where some constraints on straight alignment of structures are encounter/necessary with
the demand of resisting high temperature environments without losing their structural
integrity. They offer potential benefits in construction industry in the form of reinforced
slabs/plates, stiffened fiber-reinforced plastic super-structures, floors in bridges, ship hulls,
parallelllogram slabs in buildings and deck/skew grid of beams and girders. In aerospace
industry, swept wings of airplanes are idealized as substitute structures in the form of skew
plates. Also, complex alignment problems in bridge panels are solved by the use of plates
with skew boundary due to functional, aesthetic or structural requirements. In FGM skew
plates/shells, the skew angle is a prime key factor influencing the static, dynamic, and
buckling response of the structure. In addition, there exists strong singularity at the obtuse
vertex and hence the computation effort required for analysis of such structures increases
with increasing skew angle.
Over a period of time, the concept of functional grading is employed in sandwich

type construction, owing to their gradual and continuous variation of mechanical/thermal
properties at layer interfaces, which is not achievable in traditional sandwich arrangement.
If FGM technique is implemented in a sandwich layers, the core material could be
designed such that stiffness vary gradually from the high value at the interfaces to a lower
value at the centre, thus eliminating the large jump in material properties and hence avoids
the stress concentrations (Woodward and Kashtalyan 2011). In FGMs having sandwich
layers, modeling of plate/shell structure can be achieved in two alternative ways. In former
case, the homogenous ceramic core is introduced between the two layers of metal; thereby
the interface layer utilizes the concept of FGM. Alternatively, top and bottom skins are
occupied by the ceramic and metal isotropic materials, respectively; thereby the material in
the core portion obeys the rule of graded distribution of the constituent materials. In both
models, the gradation of material properties is governed by the volume fraction of the
individual constituent materials (Zenkour 2006). The definition of FGM could also be
employed effectively in smart structures where actuators and sensors are placed in the face
layers by appropriately selecting the core part (Xiang et al. 2010; Alibeigloo 2010; Loja et
al. 2013).
5
Thin walled structural members like plates and shells, used in reactor vessels,
turbines and other machine parts can experience large elastic deformations and finite
rotations and hence susceptible to failure due to excessive stresses induced by thermal or
combined thermo mechanical loading conditions. In such situations, analysis has to be
performed by considering geometric nonlinearity to predict the large deformation
responses. While doing so, non-linear strain part has to be taken care of for describing the
strain components. For non-linear analyses, strain part includes Green-Lagrange strain
relation by incorporating quadratic terms of in-plane and transverse displacement
components. Assumptions such as derivatives of in-plane displacement components with
respect to Cartesian co-ordinates are small and transverse displacement is independent of
thickness co-ordinate (z) are imposed on the Green-Lagrange relation to arrive von
Kármán form of strains (Fung 1965). Even though, some quadratic and cubic terms are
present in von Kármán strain part, the inadequacy is observed in the form of not capable of
defining the case of moderate rotations. When large rotations are encountered, von Kármán
strains are proven to be unsuitable to describe the actual boundary conditions, since it is
based on the undeformed co-ordinate system (Pai 2007). Hence, considering the full
geometric nonlinearity in terms of presence of quadratic terms of displacement
components seems to be vital to describe the structural response under large amplitudes.
Due to the presence of all non-linear terms, the formulation involves mathematical
complexities in the form of large matrix sizes.
During their service life, plate and shell elements are exposed to various types of
transverse and in-plane mechanical and thermal loadings. Hence, stability analysis of these
structures under such loading conditions is one of the major issues associated with the safe
and optimum design. The plate structure exhibit reserve strength after the critical load is
reached, which is generally described as post-buckling strength. It is well known
observation that the plate structures are capable of carrying additional load to a large extent
after buckling without any signs of failure. In order to fully exploit the strength of FGM
plates in carrying in-plane loads, an accurate prediction of their load resisting capacity in
post buckling region forms essential topic in this research area.
By considering the aforementioned aspects regarding analysis and behavior of

graded structures, development of an efficient and ingenious model based on accurate
numerical tool seems to be an imperative task for researchers engaged in this field. In this
6
connection, a suitable kinematics fields that incorporate realistic variation of transverse
displacement with the inclusion of bending and shear terms is necessary to accurately
predict the static and dynamic response of FGM structures, which generally exhibit the
stretching-bending phenomenon.
Further, exploiting suitable non-linear terms in the strain part is necessary to obtain
the solutions close to the practical situation for the cases involving finite strains and
moderate rotations or large deformations. Keeping the above imperative aspects in mind,
application of graded concept in sandwich plates/shells considering linear/non-linear
analysis becomes very useful for designers and researches to arrive optimum design. An
accurate modeling, simple analysis and effective design of FGM structures with the
features of skew boundary and sandwich layers based on the above criteria would certainly
serve as milestone in the field of material research.
1.2 OBJECTIVES AND SCOPE OF THE PRESENT RESEARCH

The objective of present investigation is to develop an efficient and simple 2D
model for the analysis of single/multi layer FGM structure using displacement based finite
element method. The proposed numerical approach should able to incorporate the
mechanical and thermal analyses considering the thermal-dependent properties in both
linear and non-linear sense. A FGM plate/shell with skew geometry is also to be modeled
by suitable transformation of boundary conditions from global co-ordinates into local
domain. FGM sandwich plates/shells are also to be modeled with either graded core or
homogeneous core in order to utilize the optimum material properties in an effective way.
In addition, to accurately predict the large deformation behaviour of such structures,
geometric non-linear analysis and buckling analysis beyond critical load range (i.e., post
buckling behavior) are to be performed.
Based on the aforementioned objectives, the scope of the present investigation

encompasses the following salient features.
i. To perform the linear/non-linear analysis of single/three layer (sandwich) FGM

structure with/without skew boundary considering a kinematics field that
incorporates constant/quadratic variation of thickness terms in defining the
transverse displacement (w). In-plane displacement fields (u and v) are assumed to
have cubic variations across the thickness. Such a model should be able to
7
incorporate the effect of normal strain and realistic transverse deformation in
efficient manner.
ii. In case of spatial variations of mechanical/thermal properties as in FGM, it is not
wise to ignore the heterogeneous nature of RVE (representative volume element).
In such cases, methods which consider the grading concept at both microscopic and
macroscopic level seem to be appropriate to accurately define the effective
properties. To include this aspect of FGM, Mori-Tanaka Scheme and rule of
mixture methods of homogenization are employed in the present study.
iii. Based on the proposed C0 finite element formulation and homogenization schemes,
a numerical code is initially developed in FORTRAN 90. To utilize a more
versatile numerical platform for solving different problems (e.g. sandwich FGM
plates/shells) especially where geometric non-linearity or post bucking analyses are
done, MATLAB (R2013b) tool is widely used for different analyses.
iv. Thermal analysis is done for linear static, free vibration and buckling problems by
incorporating temperature-dependent properties of the constituents. Non-linear
through-the thickness thermal distribution is assumed by virtue of graded thermal
properties of FGMs.
v. To ensure suitable assumption of core thickness having graded or isotropic material
with respect to total thickness for FGM sandwich plate/shell structures, various
schemes have been modeled that incorporate different core layer thickness with
respect to total/face sheet thickness.
vi. The developed computer codes based on FORTRAN 90 and MATLAB (R2013b)
are successfully applied to solve the static (linear/ geometric non-linear), dynamic
(free/forced response) and buckling/post buckling responses of FGM plate/shell
structures having single/sandwich layers with/without skew boundary. While
solving the problems, different choices of ceramic and metal constituents are
considered. Several parameters such as aspect ratio, side-thickness ratio, radius-
thickness ratio, skew angle, boundary conditions, shell curvatures, and volume
fraction index are considered to show their influences.
vii. Comprehensive numerical results are presented in the form of tables and graphs to
show different responses of single/multi layer FGM structures, which should be
useful for researchers/engineers working in the field.
8
1.3 ORGANIZATION OF THESIS
A brief overview and various issues that serve as motivation for the present investigation
are summarized in Chapter 1. The objectives and scope of the present research work are
also described at the end of Chapter 1.
Chapter 2 presents the brief overview of existing literatures that serve as
background for the present research. In the first part, various shear deformation theories
available for the static, free/forced vibration and buckling analyses are discussed along
with the merits of each theory. Subsequently, the research works related to the analysis of
composite/FGM skew plates under mechanical and thermal loadings are discussed. In
addition, the works related to analysis of FGM sandwich plates are elaborated along with
their interesting findings. In the end, non-linear bending and post buckling studies
performed on laminated composites/FGM structures are outlined.
Chapter 3 explains the brief mathematical formulation based on constant and
quadratic variation of transverse displacement in the kinematics field. The formulation that
assumes constant transverse displacement is based on Reddy’s higher order shear
deformation theory with the implementation of C0 isoparametric formulation. The first part
covers the governing equations for linear static, vibration, dynamic and buckling analysis
followed by non-linear formulation for static and post buckling analyses. Solution
techniques adopted for different analyses combined with computer coding are included at
the end of the chapter.
Chapter 4 demonstrates the application of the present formulation in solving
various numerical problems related to functionally graded plates/shells under different
loading conditions. In each case, the applicability of the developed coding based on C 0
finite element formulation has been ensured by comparing the present results with the
results published in literatures by performing the convergence study. Numerical examples
of FGM skew plates/shells are presented for different linear analyses by varying different
parameters such as aspect ratio, thickness ratio, curvature ratio, boundary conditions, skew
angle and volume fraction parameter. In case of sandwich arrangement, the influence of
material grading on thickness range of bottom/core/top layers of plate/shell are studied
under two different types of modeling. Finally, non-linear analysis is performed for
bending and post buckling of FGM plates/shells. Several new results are presented for
linear and non-linear analyses of FGM skew plates/shells considering single/sandwich
configurations.
9
Chapter 5 summarizes different conclusions and observations of the present
research work in a concise form. Lastly, the scope of the future works in context to the
present research is described followed by the list of publications in Journals and
conferences accomplished from the present research work.
10
CHAPTER 2
LITERATURE REVIEW
2.1 INTRODUCTION
FGM structures are highly inhomogeneous, by virtue of choice of material constituents. In
addition if skew alignment is encountered in such structures, the analysis becomes more
tedious and hence proper attention has to be paid to ensure the reliable and optimum
design. In this context, an efficient and simple 2D modeling of FGM structures based on
accurate analytical/numerical technique is essential which should also be able to predict
the results as accurate as 3D elasticity solutions. Further, if sandwich layers are
incorporated in FGM configuration proper modeling of core and face sheet layers relies on
suitable choice of material constituent. More often these structures undergo large
amplitudes with moderate rotations. In such cases, sufficient knowledge should be
acquired regarding the effect of geometric non linearity and buckling response beyond
critical load point. To solve the aforementioned issues, different analytical/semi
analytical/numerical solutions are proposed by many investigators in the previous
literatures by considering appropriate displacement field and homogenization scheme.
Since this research area is relatively new and gaining more attention from
researches, a vast body of research summary exists in the literature for static, dynamic and
buckling analyses. Hence by keeping the aim and objective of the present investigation in
mind, only the literatures that are related to the current research topic are reviewed in this
chapter. Meanwhile, care has been taken to include the recent literatures that are related to
this topic. Since limited numbers of literatures are available regarding FGM
skew/sandwich plates/shells under large deformation analysis, some of the literatures
related to composite skew/sandwich plates/shells are also appraised. Before proceeding to
the present research problem, a critical survey and assessment of the existing literatures
have been performed which are categorized into the following sections.
 Brief review of various linear/non-linear shear deformation theories of
plates/shells
 Static, free vibration, buckling and dynamic response of FGM plates/shells
 Static, free vibration, buckling and dynamic response of
o Laminated composite skew plates/shells
11
o FGM skew plates/shells
 Static, free vibration and buckling response of
o Laminated composites sandwich plates/shells
o FGM sandwich plates/shells
 Geometric non-linear analysis of
o Laminated composite plates/shells
o FGM plates/shells
 Post buckling responses of
o Laminated composite plates/shells
o FGM plates/shells
In addition to above sections, a separate sub section is also included at the end where the
important observations obtained from previous literatures are discussed.
2.2 BRIEF REVIEW OF VARIOUS SHEAR DEFORMATION

THEORIES OF PLATES/SHELLS
This section is separated into four sub sections to provide a brief outline about the
various shear deformation theories developed for laminated composite plates/shells. At
first, various shear deformation theories developed for the analysis of laminated composite
plate are discussed followed by the implementation of such theories for linear/non-linear
analysis of functionally graded plates in section two. While third section comprises a
detailed discussion on linear/non-linear shell theories developed for analysis of laminated
composites and the section ends by providing necessary information about application of
shell theories to analyze functionally graded shells. The reference order is not intended to
imply priority of any particular theory.
A solution accuracy of any analysis problem largely depends on the assumed
displacement field based on which the strain equations are formulated. In order to capture
the accurate profile of shear deformation, many theories were proposed and implemented
to analyze the static and dynamic response of structures under complex loading conditions.
Initially, two plate theories are discussed which are widely adopted in most of the
literatures to model the plate geometry. The first theory does not consider the effect of
transverse shear deformation; while the second theory accounts for it. In both the theories,
normal stress in the thickness direction was assumed as zero. In 1850, the first theory is
emerged and named as thin-plate theory or Kirchoff’s theory. The second theory is
popularly known as Mindlin, Reissner-Mindlin and Mindlin-Reissner theory (1944; 1945a;
12
1947). The initial credit goes to Ashton and Whitney (1970) for implementing CPT for the
analysis of composite structures. Linear and constant variation of in-plane and shear strain,
respectively, is assumed in the investigation. The main drawback of the theory comes in
the form of neglecting the transverse shear contribution in the deformation process. As a
further development, displacement theory that includes the effect of rotary inertia and
shear is formulated by Mindlin (1951) for elastic isotropic plates. Later on, the Mindlin
theory (1951) was extended to laminated anisotropic plates (Yang et al. 1966; Whitney and
Leissa 1969; Whitney and Pagano 1970), named as first order shear deformation theory
(FSDT) and provides a balance between computational efficiency and accuracy at global
domain and suitable for thin and moderately thick laminates plates. But the theory fails to
predict the responses at local domain, for example, the interlaminar stress distribution
between layers, delaminations, etc. Since the theory consider the transverse shear strain in
constant sense, use of arbitrary number known as shear correction factor is mandatory in
the analysis (Pandit 2009).
It has been observed in many literatures that the performance of the FSDT is
primarily dependent on these shear correction which again fluctuate according to
geometry, boundary and loading conditions (Liu et al. 2007; Reddy 2007). To offer the
solution for the various difficulties encountered in the above theories (CPT and FSDT),
higher power of thickness term is incorporated in in-plane displacement fields to describe
the actual warping of cross sections, accurately. The benefit of elimination of shear
correction factor recommended by HSDT by incorporating realistic variation of transverse
shear deformation made them to be employed in various analyses, thereafter. In developing
HSDT for analysis of plate structures, a significant contribution was recorded by
Hildebrand et al. (1949) and Basset (1890). As further improvement in this direction, Lo et
al. (1977a, b) proposed higher order plate theory incorporating non-linear distribution of
in-plane displacement with respect to thickness co-ordinate for homogenous and laminated
plates. The theory incorporates four terms in in-plane part in addition to three terms for
transverse displacement and thus finally leads to eleven unknowns in the kinematics
model.
As a continued effort, different higher order models were proposed in the literature
to account for realistic variation of transverse shear stresses and strains (Nelson and Larch
1974; Cho et al 2007; Makhecha et al. 2001) and highlighted the importance of higher
order terms in predicting the accurate structural responses. To further refine HSDT, cubic
and linear variation for in-plane and transverse displacement, respectively, is considered in
13
the Taylor’s expression by Kant et al. (1982) to analyze the thin and thick composite
plates. But the disadvantage lies with the higher number of unknowns exists in the theory
compared to FSDT. To solve this issue, a simple higher order theory is proposed by Reddy
(1984b) for analysis of laminated plates. The proposed theory has cubic variation for in-
plane part, while constant variation was assumed for transverse displacement component.
The various higher order unknowns exist in the kinematics field are solved by the
satisfaction of free boundary condition of transverse stresses on the upper and lower
surfaces. While doing so, the number of unknowns are considerably reduced which
obviously reduce the computational effort required for the analysis. In addition to above
theories, a 3D continuum-based theory is also employed to predict the interlaminar stresses
in a composite laminates, but the computational cost is a major concern of this theory.
To evaluate the transverse stresses in composite and sandwich laminates, a set of
higher order theories were proposed by Kant and Manjunatha (1994) by employing C0 iso-
parametric finite elements. For analysis of hybrid/smart graded plates, a higher order shear
and normal deformation theory is employed by Shiyekar and Kant (2010). The electric
field is approximated as layer wise linear model through the thickness of the piezoelectric
fibre reinforced composites, while displacement function is approximated by Taylor’s
series expansion. Further, a refined higher order model is proposed by Swaminathan and
Patil (2008) to solve the natural frequency of simply supported anti-symmetric angle-ply
composites and sandwich plates. Some of the other higher order theories that identify
seven unknowns (Kant and Pandya 1988), nine unknowns (Ren 1986; Pandya and Kant
1988b; Pradyumna and Bandyopadhyay 2008), twelve unknowns (Jha et al. 2013c) in the
assumed kinematic model can also be located in the literature. While few other available
higher-order theories propose equivalent number of unknowns as in FSDT e.g., third-order
shear deformation theory (Reddy 1984b; Reddy 2000), sinusoidal shear deformation theory
(Touratier 1991; Ferreira et al 2005b; Zenkour 2006), hyperbolic shear deformation theory
(Soldatos 1992; Xiang et al. 2009; Akavsi 2010; Grover et al. 2013), exponential shear
deformation theory (Karama et al. 2003), and trigonometric shear deformation theory
(Mantari et al. 2012a), but their equations of motion are more complicated compared to
FSDT.
A comprehensive document that list different shear deformation theories with
emphasis on estimation of transverse/inters laminar stresses in laminated composites was
reviewed by Kant and Swaminathan (2000). As a parallel effort, a review of refined
displacement and stress-based shear deformation theories for analysis of isotropic and
14
anisotropic laminated plate was compiled by Ghugal and Shimpi (2002). In addition, a
collection of different laminated theories based on displacement hypothesis was submitted
by Liu and Li (1996), which include shear deformation theories, layer wise theories, zigzag
theories, and global-local double-superposition theories. Since FGM layers do not
introduce any interfaces in their geometry, due to their smooth spatial variation of effective
properties, further discussion about zig-zag theory seems to be inappropriate at this point,
and hence subsequent discussion has been continued to discuss implementation of various
shear deformation theories for analysis of FGM plates.
The CPT which neglects the effect of shear deformation has been extended to
perform the stability analysis of functionally graded plates (Javaheri and Eslami 2002;
Zhang and Zhou 2008; Mohammadi et al. 2010; Bodagi and Saidi 2011). In some research
works, the FSDT has been employed for free vibration and buckling analysis of
functionally graded plates by many researchers (Croce and Venini 2004; Ganapathi et al.,
2006; Zhao and Liew 2009a; Hashemi et al. 2010; Hashemi et al. 2011). Batra and Jin
(2005) employed FSDT to analyze free vibration problem of FGM plates in combination
with finite element method. Few of the earlier works include quadratic, cubic and higher
order variation of in-plane displacements through the thickness of the plate (Reddy 2000;
Karama et al. 2003; Zenkour 2005a; Zenkour 2005b; Xiao et al. 2007; Matsunaga 2008;
Pradyumna and. Bandyopadhyay 2008; Fares et al. 2009; Talha and Singh 2010;
Benyoucef et al. 2010; Atmane et al. 2010; Talha and Singh 2011; Meiche et al. 2011;
Mantari et al. 2012b) with the combination of either constant/linear/quadratic variation of
transverse displacement component. A sinusoidal shear deformation theory (SSDT) having
four unknowns that accounts for sinusoidal variation of transverse shear stresses and have
resemblance with conventional sinusoidal shear deformation theory is proposed for
bending, vibration and buckling analysis of functionally graded plates by Thai and Vo
(2013). Similarities between the conventional and improved SSDT are observed by means
of equations of motion, boundary conditions and stress resultant expressions.
A Reissner’s mixed variation theorem (RMVT) for bending analysis of functionally
graded plates has been formulated by Brischetto and Carrera (2010) that incorporate both
displacements and transverse normal/shear stresses as primary field variables in order to
obtain the significant enhancement over classical models based on principal of virtual
displacements (PVD), where only the displacements are assumed as primary variables.
Different orders of expansion have been considered for primary variables through the
thickness, and these unknowns are described either by single or layer wise theory. Such a
15
theory enables an analyst to combine the different plate cases in a unified manner. Qian et
al., (2004) obtained solutions for static, free vibration and forces response of thick FGM
plates using higher order shear and normal deformation plate theory. A generalized shear
deformation theory proposed by Zenkour for the analysis of cross ply laminated and visco
elastic composite plates (2004a; 2004b; 2004c) has been extended to static analysis of
functionally graded plates (Zenkour 2006) which enforces traction-free boundary
conditions at the faces of the plate. The theory proposed by Zenkour (2006) has similarity
with the higher order theory of Reddy (2000) and have similar unknowns as involved in
FSDT, and also able to predict the transverse stresses in accurate manner.
Different forms of shear deformation theories that include higher order terms in
HSDT for functionally graded plates (Mantari et al. 2012b), trigonometric HSDT for
exponentially graded plates (Mantari and Soares 2012 a), including thickness stretching
effect in HSDT for functionally graded plates (Mantari and Soares 2013 a), hybrid quasi
3D shear deformation theory (Mantari and Soares 2012 b) and generalized HSDT (Mantari
and Soares 2013 b) for static analysis of advanced composite plates, has been proposed by
Mantari and his associates. An optimized sinusoidal HSDT incorporating sine and cosine
terms in in-plane and transverse displacement, respectively, is proposed by Mantari and
Soares (2014), recently, for the bending analysis of functionally graded plates and shells.
The proposed SSDT include the effect of thickness stretching by means of shear strain
shape functions and depends on arbitrary parameters m and n (these parameters are
selected based upon appropriate displacements and stresses). The theory employs no shear
correction factor since it satisfies the tangential stress-free boundary conditions on the
plate boundary surface. To predict the accurate evaluation of mechanical stresses in
functionally graded plates and shells, thickness stretching effect has been incorporated in
the formulation by Carrera et al. (2011a). A quasi 3D hybrid theory also known as
polynomial and trigonometric theory is proposed by Ferreira and his co-workers (Neves et
al. 2011; Neves et al. 2012b; Ferreira et al. 2011a) for static and free vibration analysis of
functionally graded plates in the framework of mesh free methods. Having a brief
overview about various shear deformation theories, the application of such theories for the
non-linear problems of functionally graded plates is discussed in the following paragraph.
The CPT is implemented to study the non-linear cylindrical bending of FGM plates
with the variation of material properties as a sigmoid function in the thickness direction
(Kaci and Bakhti, 2013). To perform the non-linear thermo-elastic response of FGM plates
(Praveen and Reddy 1998; Zhao and Liew 2009b) FSDT is incorporated with the constant
16
variation of transverse shear stress through the thickness. Recently, FSDT is employed to
study the geometric non-linear analysis of functionally graded plates using cell-based
smoothed three-node Mindlin element (Van et al. 2014). To deal with small strains and
moderate rotations, von Kármán assumptions are imposed by adopting C0 HSDT
formulation. In the study, moderate strain gradient elasticity theory and von Kármán
assumptions are implemented. Shen (2002) employed Reddy’s theory for non-linear
bending of FGM plates subjected to transverse uniform and sinusoidal load. Similarly,
third order plate theory of Reddy and von Kármán assumptions are incorporated for
kinematics and kinetics field to obtain the non-linear thermo-elastic bending response of
FGM plates by Aliga and Reddy (2004). To analyze the non-linear cylindrical bending of
FGM plates under thermal and mechanical loads, a four variable refined plate theory is
proposed by Fahsi et al. (2012). The transverse displacement consists of bending and shear
component; where, bending components do not contribute toward shear forces and vice
versa. The theory accounts for quadratic variation of transverse shear strains across the
thickness and eliminates the use of shear factor by incorporating zero traction conditions
on the top and bottom surfaces of the plate. The theory proposed by Fahsi et al. (2012) is
extended to study the cylindrical bending of FGM nano composite plates by Bakhti et al.
(2013) in conjunction with Von Kármán theory and potential energy principle. In a parallel
track, Kaci et al. (2013) proposed cylindrical bending of FGM nano composite plates
reinforced by single walled carbon nano tubes by incorporating Reddy’s third-order plate
theory and von Kármán geometric nonlinearity to describe kinematic and kinetic fields. A
general nonlinear third-order plate theory that accounts for (a) geometric nonlinearity, (b)
microstructure-dependent size effects, and (c) two-constituent material variation through
the plate thickness (i.e., functionally graded material plates) is presented using the
principle of virtual displacements by Reddy and Kim (2012). The proposed couple stress
theory includes a material length scale parameter that can capture the size effect of
functionally graded material, in exact sense.
A catalog of non-linear classical theories is proposed for thin shells based on
Kirchhoff-Love hypotheses (Donnell 1934; Novozhilov 1953; Sanders 1963; Koiter 1966;
and Ginsberg 1973) by including shallow shell assumptions. In Donnel’s theory (Donnell
1934), infinitesimal in-plane displacements are considered; while transverse displacement
is assumed to be in the order of shell thickness. Since the theory discounts for the in-plane
inertia, it gives accurate results for only thin shell categories which is quite contradictory
observation compared to Donnel’s linear shell theory (1933). In the theory (Donnell 1934),
17
non-linear terms are retained only in the transverse displacement and neglected for in-
plane field; which is analogous to the von Kármán assumptions assumed in non-linear
plate theories. The classical shell theory (CST) proposed by Sanders (1963) is considered
to be a more refined form of tensorial based shell theory. An improved form of Sander’s
theory is presented by Koiter (1966) in the name of Sander-Koiter theory to consider finite
deformations with small strains and moderate rotations. The non-linear terms appear in
strain-displacement relations are dependent on both in-plane and transverse displacement
components. The consequence of curvature changes and torsion of the middle-surface are
assumed in the linear sense by both the theories (Sanders-Koiter 1966; Donnell 1934).
However, Donnell’s theory (1934) yield accurate results for moderately thick shells and
modes of high circumferential wave number, only.
In the non-linear shell theories proposed by Novozhilov (1953) and Ginsberg
(1973), non-linear terms are added to the curvature and torsion part. However, the strain-
displacement relation identical to that of Sander-Koiter theory (Koiter 1966) is assumed.
Because the shear deformation and rotary inertia are neglected in the above classical
theories, various shear deformation theories are developed in due course of time as an
alternative solution. In the category of shear deformation theories, the dominant role is
played by two theories namely; first-order and higher-order shear deformation theories;
while the first category demand the suitable estimation of shear factor to satisfy
equilibrium condition and the later theory employs the boundary condition similar to
higher order plate theory. Parisch (1995) and Sansour (1995) proposed shell theories to
introduce quadratic assumption of shell displacement over the shell thickness. The linear
shell theory that incorporate the effect of thickness stretching was submitted by Carrera et
al. (2011a) and Ferreira et al. (2011b). An enhanced form of first order shear deformation
theory in the frame work of finite element formulation is presented by Arciniega and
Reddy (2007a) for the non-linear analysis of ample range of shell geometries that include
isotropic, laminated composite and FGM structures.
By incorporating Sanders-Koiter non-linear terms (Koiter 1966), Reddy and
Chandrashekhara (1985) developed non-linear FSDT, by defining the deformation process
in terms of five independent variables (three translations and two rotations). As a further
improvement in this direction, non-linear terms are included in Reddy’s theory and
implemented in the framework of finite element method (Reddy 2004; Dennis and
Palazotto 1990; Palazotto and Dennis 1992). An extended application of higher order shell
theory to the case of anisotropic sandwich shells having compressible core is evident from
18
the work of Hohe and Librescu (2003). They assumed the Kirchhoff-Love hypotheses for
the face sheets and a second/third-order power series expansion for core displacements. As
a development, Reddy in association with Amabili developed a refined non-linear shell
theory for closed and open shells by retaining rotary inertia, shear deformation and non-
linear terms in both in-plane and transverse displacements (Amabili and Reddy 2010). The
so formed new theory (Amabili and Reddy 2010) has proved excellent performance in
predicting the large-amplitude vibrations of moderately thick laminated circular cylindrical
and deep shells (Amabili 2011) and curved panels (Alijani and Amabili 2013). As a further
progress, modification of the theory in the form of incorporating thickness stretching effect
and geometric imperfections is performed by Amabili (2013) by means of third-order
variation of normal strain in the non-linear theory developed by Amabili and Reddy
(2010). More recently, a theory that accounts for normal strain in the kinematics field by
means of third-order variation of thickness is executed by Amabili (2014).The benefit of
retaining transverse normal strain components is that it utilizes all the constitutive
equations and such a consideration is predominantly suitable for materials where large
deformations are achieved by large thickness reduction.
In addition to above shell theories, a huge list of tensor based geometrically non-
linear shell theories are also proposed in the literature (Eremeyev and Pietraszkiewicz
2004; Opoka and Pietraszkiewicz 2004; Pietraszkiewicz and Szymczak 2005; Arciniega
and Reddy 2007b; Opoka and Pietraszkiewicz 2009; Berdichevsky 2010; Xiaoqin et al.
2010; Pietraszkiewicz 2012; Steigmann 2013). A widespread assemblage of various
higher deformation theories is carried out by Reddy (2004), Amabili (2008), and Carrera et
al. (2011b). In addition, a profound discussion is performed on linear shear deformable and
zigzag theories by Reddy and Arciniega (2004) and Carrera (2002; 2003). More recently,
an in depth review of various shell theories is executed by Alijani and Amabili (2014) and
they also made an attempt to discuss several other related aspects of non-linear vibration of
shells, for example, fluid-structure interaction, geometric imperfections, influence of
thermal and electrical loads in a brief manner. In what follows, a discussion regarding
application of various shell theories to FGMs is discussed.
Based on Love’s shell theory (Love 1952), Loy et al. (1999) studied the frequencies
of simply supported FGM cylindrical shells using Ritz method. This study was further
extended to incorporate the effects of various boundary conditions on natural frequencies
of FGM cylindrical shell by Pradhan et al. (2000). Based on TSDT of Loy et al. (1999),
Najafizadeh and Isvandzibaei (2007) presented the free vibration response of thin
19
cylindrical shells having arbitrarily fixed ring support along the shell. This study was
further extended by the authors to study the influence of various shear deformation theories
(Najafizadeh, 2009) on free vibration response of cylindrical shells. The FSDT as a special
case of higher order shear deformation theory of Reddy is assumed to represent the
kinematics field. The FSDT considering rotary inertia and transverse shear strains is used
to study the effect of thermal load on free vibration, buckling, and dynamic stability of
FGM shells by Sheng (2008). Based on FSDT, free vibration of cylindrical, conical, and
annular FGM shell structures is studied by Tornabene and his co-workers using four-
parameter power law distribution ( Tornabene 2009a, Tornaence 2009 b). Large
deformation vibration behavior of FGM cylindrical shell of finite length embedded in
elastic medium is under thermal environment is presented by Shen (2012). Higher order
shear deformation theory that includes the effect of shell-foundation interaction is
incorporated in the study. General shell theory combined with finite element method is
employed to study the vibration analysis of FGM cylindrical, doubly curved, hyperbolic
paraboloid shell (Yang 2012). More recently, Ebrahimi and Najafizadeh (2014) studied the
free vibration response of two dimensional functionally graded (2D FG) cylindrical shells
using Love’s first approximation CST.
The problem of geometric non-linearity, initial geometrical imperfection and
Pasternak type elastic foundation based on CPT is solved for non-linear axi-symmetric
response of shallow spherical FGM shells under thermal and mechanical loads by Duc et
al. (2014). Based on FSDT, finite element method has been employed to study the dynamic
stability of functionally graded shallow spherical shells (Ganapathi 2007). Geometric non-
linearity is considered in von Kármán sense and Newton iteration schemes are considered
to solve non-linear iteration equations. Non-linear axi-symmetric dynamic buckling
behavior of clamped FGM spherical caps is performed by Prakash et al. (2007) based on
FSDT and von Kármán assumptions. Bisch and his co-workers performed static and
dynamic non-linear analysis of FGM spherical shells under different loading environments
by considering CST and geometric imperfections (Bich 2009; Bich et al. 2010; Bich et al.
2011; Bich et al. 2012).
Hence an exact kinematic model that incorporates the realistic variation of shear
distribution through the thickness and considers the effect of normal strain in the transverse
direction seems to be very important for accurate modeling of FGM structures under
linear/non-linear responses.
20
2.3 STATIC, DYNAMIC AND BUCKLING RESPONSES OF FGM
PLATES/SHELLS
3D elasticity solutions (Pagano 1969; Pagano 1970; Srinivas and Rao 1970;
Srinivas et al. 1970) are generally utilized to assess the accuracy of various 2D plate
theories (Pandya and Kant 1988a; Pandya and Kant 1988b; Pandya and Kant 1988c; Reddy
2004). In this regard, several bench mark solutions are placed in the literature based on 3D
theories for simply supported laminated plates. But the solution methodology lays the
limitation for FGMs, where the material properties are generally inhomogeneous in nature.
Therefore many displacement based 2D theories are proposed in conjunction with
analytical, semi-analytical and numerical solutions. An exceptional introduction to the
fundamentals of FGMs and a comprehensive literature review in FGM technology was
provided by Suresh and Mortensen (1998). In addition, Birman and Byrd (2007) have
documented an exhaustive list of research works regarding developments in FGM research
by addressing the various topics like characterization, modeling and analysis of FGM.
Important discussion include manufacturing, design, homogenization of particulate FGM,
heat transfer problems, stress, stability and dynamic analyses, fracture studies and various
application areas. Recently, Jha et al. (2012a) have made an attempt to present the
exhaustive literature survey on deformation, stress, vibration and stability problems of
FGM plates. In this section, various available literatures on static, dynamic and buckling
analyses of FGM plates/shells published since 1999 are briefed. Although it was an
unfeasible task to discuss all the works in a single document, an endeavor has been made
by the author to incorporate the important and relevant works available in this area.
The free vibration response of FG cylindrical shells made of stainless steel-nickel
alloy was studied by Loy et al. (1999) based on Love’s shell theory (Love 1952) and
Rayleigh-Ritz method. The effective mechanical properties of the shell are assumed to be
graded in the thickness direction and follow power law distribution in terms of volume
fraction of constituents. Based on the assumption of nickel/stainless steel on inner/outer
surfaces, two types of FGM shells (type I and type II) are analyzed in the study. For the
value of circumferential wave number greater than one, natural frequencies of type II FG
cylindrical shells (nickel on outer surface and stainless steel on inner surface) are recorded
to be higher than type I FG (nickel on inner surface and stainless steel on outer surface)
cylindrical shells and for circumferential wave number other than this value, FGM shells
exhibit quite opposite trend. For type I and type II FG cylindrical shells, for all the values
21
of volume fraction indexes, frequency values between stainless steel and nickel alloy are
discerned.
A finite element model of cylinder was employed to study the thermo-elastic
response of FGMs by Praveen et al. (1999). Considering the fact that the inner surface of
the cylinder was subjected to rapid temperature rise, the solution for unsteady heat transfer
equation has been solved. The constitutive equation has been updated at each time step
which is further solved by energy equation. The inertia terms and thermo-mechanical
coupling are neglected in the equilibrium equations. When the average volume fraction of
the ceramic attains value less than 0.5, no significant change in maximum temperature was
observed. This reflects the statement that FGMs with 50% of volume fraction of ceramic
are good enough to withstand large temperatures, and no gain has been visualized beyond
that range. Variation of radial compressive, radial tensile and hoop stresses is plotted
against average volume fraction of ceramics of FGM cylinder. When temperature-
dependent properties are considered, the maximum tensile stress at the inner surface is
independent of the average volume fraction of ceramic below 50%. Thus increasing the
volume fraction reduces the tendency of any crack growth at tensile mode. The final
observation was that, beyond certain percentage of average volume fraction of ceramic, an
asymptotic response has been observed. Again this response is a function of different field
variables chosen to perform the study.
The free vibration study performed on FG cylindrical shells by Loy et al. (1999)
has been extended to study the influence of boundary conditions on free vibration response
of FG cylindrical shells by Pradhan et al. (2000). The displacement field and solution
method similar to that of Loy et al. (1999) was considered. Clamped, simply supported and
free boundary conditions are considered to generate the frequency results. Frequency
characteristics similar to that of isotropic case are discerned for FG shells also. Further,
they observed that for smaller value of volume fraction index (n=0.1), the frequency data
of FG cylindrical shells are close to isotropic metal shell (stainless steel shell). For other
higher values of volume fraction index (n>>10), the frequency response of FG cylindrical
shells are close to that of isotropic ceramic shell (zirconia shell). This observation guides to
choose the appropriate value of volume fraction to get the required frequency response of
FG shells. For different L/R (length/curvature) ratios, the frequency responses of clamped
and free edge cylindrical shells are found to be identical.
Initially stressed FGM plates are analyzed for free and forced vibration response by
Yang and Shen (2002) under thermal environment. Temperature dependent material
22
properties and having power law variation through-the thickness are considered. Reddy’s
higher order shear deformation theory was assumed and uniform temperature variation
under thermal part was considered. Modal superposition method has been employed to get
the transient response of the plate under lateral dynamic loads. It was observed that for the
mixed material mixture, frequency rises by elevating in-plane tension but reduces by
increasing initial edge compression. Since Young’s modulus decreases at higher
temperatures, the frequency parameter declines at higher temperatures. Also, higher
bending moments compared to isotropic plates are observed in FGM case, due to the lower
thermal expansion of the plate at top compared to bottom. It has been noticed that dynamic
response of FGM plates are not necessarily lie between isotropic cases under thermal
loading conditions.
The axi-symmetric stability of circular FGM plates was considered by Najafizadeh
and Eslami (2002) based on Love-Kirchoff hypothesis. The linear Sander’s shell theory
was employed to approximate the strain displacement field. Results are presented for
simply supported and clamped boundary conditions and material properties are assumed to
be graded in the radial direction. They observed that the critical buckling load of FG plates
was lower than corresponding pure isotropic circular plates.
A three-dimensional analytical solution was proposed by Vel and Batra (2003) for
simply supported functionally graded plates subjected to time-dependent thermal loads.
The governing transient heat conduction equation has been reduced into ordinary partial
differential equations by means of Laplace transform equations and then solved by the
power series method. The micromechanical models based on Mori-Tanaka and self
consistent scheme are incorporated in the study. The important finding of the study was
that, the transient longitudinal stresses are approximately 8 times their respective steady
state value, when rapid time-dependent surface temperatures are prescribed. But the
transient stresses are less than their steady state values. As the time elapses, the pattern of
stress changes from compressive to tensile in case of both longitudinal and transverse shear
stresses.
The study carried out by Vel and Batra (2003) has been extended to study the
vibration of functionally graded rectangular plates by Vel and Batra (2004). The solution
method and micromechanical models identical to that of Vel and Batra (2003) were
incorporated in the model. In addition, the transient response of the plate was performed
under the sinusoidal spatial distribution of pressure applied on the top surface. The authors
manifested that, the displacement and stresses in the functionally graded plates exhibit anti
23
symmetric pattern with respect to the mid-plane, by virtue of their anti symmetric
properties about the mid-plane. Under forced response, the displacements and stresses are
large as the forcing frequency approaches the natural frequency. Also, the normal and
transverse stresses are more for the case of sinusoidal normal pressure applied on the top of
the plate. Even though, the transverse normal and shear stresses are computed by
integrating the 3D elasticity equations, a considerable deviation between the analytical
solution and the CPT results was noticed, especially for thick plates. The FSDT results are
found to be close to analytical solution than TSDT results.
In addition to above discussed works, Batra and his co-workers studied the fracture
concepts in functionally graded materials (Jin and Batra 1996), stress intensity relaxation
study in the cracked functionally graded material subjected to thermal shock (Jin and Batra
1996), and R-curve and strength behavior of functionally graded materials (1998). Since
these topics are not relevant to present research topic, not discussed briefly in this section.
Najafizadeh and Heydari (2004) studied the thermal buckling of functionally
graded circular plates based on HSDT (Reddy and Khdeir 1989). The fundamental partial
differential equations are established by variational approach and mechanical properties
are assumed to be graded in the thickness direction in proportion with their volume
fraction index. They compared the various numerical results with the FSDT and CPT
results and concluded that the CPT and FSDT overestimate the buckling temperature
parameter. Under uniform temperature rise, buckling temperature was found to be lower at
volume fraction index equals 2.7, while maximum at volume fraction index equals 10, and
thus it was recommended to opt for the value between 4.0 and 10.0 for the parameter.
Further, the critical buckling temperature of FGP was observed to decline by reducing the
value of volume fraction index.
Free/forced vibration and static study of thick functionally graded plates was
performed by Qian et al. (2004). The meshless Petrov-Galerkin method in the framework
of higher order shear and normal deformable plate theory was utilized to perform the
analysis. In addition, the response of the plate under impulse load was considered by
employing different values of volume fraction index. To account for the interaction
between the adjacent inclusions, Mori-Tanaka approach was used to derive the effective
elastic constants. The top surface of the plate was loaded with sinusoidal form of traction
represented as Fourier series to obtain the static solution and the solution was obtained by
the superposition method. A time dependent harmonic normal pressure of uniformly
distributed was applied for the time period 0≤t≤5 ms and suddenly removed. Through-the-
24
thickness variation of deflection for both isotropic and FGM plate exhibit un symmetric
variation about the mid-surface, due to the non symmetric pattern of applied load. Also,
the tensile stresses observed at the bottom segment of the plate are reduced towards top by
the addition of ceramic content and compressive stresses at the top are increased with the
addition of ceramic part. As far as the effect of volume fraction was concerned, the
deflection variation with volume fraction index was noticeable, while it is not for the case
of axial stress variation of FG plates. Also for linear volume fraction value, the natural
frequencies of pure ceramic and pure metal are the upper and lower bounds of the
frequencies of the FG plate. Regarding dynamic response, the oscillation time period of
pure ceramic plates is found to be lesser than that of pure metallic plate.
Ferreira et al. (2005a) presented the static problem of functionally graded plates
using third-order shear deformation theory in conjunction with meshless method. Two
kinds of homogenization schemes namely, Mori-Tanaka approach and rule of mixture are
adopted to estimate the effective properties of the FG plate. Two types of material
combinations are considered; the one with equal Poisson’s ratio value of constituents and
the other combination with a wide variation of Poisson’s ratio. An interesting observation
regarding the influence of Poisson’s ratio of two constituents on displacement has been
noticed i.e., when Poisson’s ratio becomes equal, both the models produce quite close
results and have large variation for different values of Poisson’s ratio. While plotting the
axial stress profile for FG plates, lower and higher values of volume fraction index have
sharp gradient change near the bottom and top, respectively, due to sharp variation of
material properties.
A micro mechanical model based on Mori-Tanaka scheme, global collocation
method, the FSDT and HSDT are employed by Ferreira et al. (2006a) to study the natural
frequencies of FGPs. The solution proposed in the work does not require any nodal
connectivity procedure and evaluation of integral was performed over a sub domain.
Different parameters such as boundary conditions, thickness ratio, and volume fraction
index are considered to present the natural frequencies of FGM plate. It was noticed that
frequency depends on number of collocation points, their corresponding locations, and the
parameter c present in the multi quadratic basis functions.
The proportionality constant that exists between the homogenous and FGM plates
was derived and proposed by Abrate (2006) for static, free vibration and buckling
problems. The author observed that the natural frequencies, deflection and buckling load of
FGM plates are proportional to those of corresponding homogeneous isotropic plate, while
25
the other parameters are kept constant. Different examples are selected from the literature
to show the correlation between the homogenous and FGM plates for different analyses.
Numerical expressions are derived that will supply the suitable proportionality constant to
predict the static, vibration and buckling responses of FGM plates.
In Part I submitted by Chi and Chung (2006), FG plates are analyzed under
mechanical loading and series solutions for different kinds of plates based on various
micromechanical models are proposed. Extensive numerical problems are performed
based on the solutions presented in Part I by Chi and Chung (2006) as Part II. The graded
properties of the FGM are described by power-law, sigmoid and exponential functions.
The CPT and Fourier series are employed in the study to define the displacement field and
closed form solutions, respectively. They concluded that the location of the neutral surface
of the FGM plates depends on the ratio E1/E2 for particular material distribution or
variation of material properties in the thickness direction, while it was independent of the
aspect ratio or the external loads. The stresses in the FGM plates are not linearly
proportional to z and function of the product z. E(z). The tensile stresses are observed to
be maximum at the bottom of plate, while the location of maximum compressive stress
moves towards the inner side, rather than at the top surface of the plate.
A 3D based solution was proposed by Uymaz and Aydogdu (2007) for vibration
analysis of functionally graded plates based on the small strain linear elasticity theory. The
Chebysheve displacement functions combined with Ritz method was employed to solve
the vibration problem of FGM plates. Prominence has been given to sketch the influence of
various boundary conditions on frequency response of FG plates. For all the boundary
conditions, rise in a/h ratio increases the frequency parameter and converges at a/h=50.
Also, constant density with variable Young’s modulus produce lower frequency values
compared to the case, where both Young’s modulus and density are treated as variables.
Ang (1988) considered an anisotropic elastic slab containing a plane crack under
impact loading. The crack under consideration is assumed to be perpendicular to the
boundary of the slab. The problem is condensed to a Fredholm integral equations of the
second type after which the stress intensity factors are determined. Ang and Gurnel (1996)
evalueated the stress field around an arbitrary number of arbitrarily-located planar cracks
in an anisotropic half-space. The antiplane deformation of multilayered medium is taken to
perform the analysis. It is considered that the strip is made up of numerous layers of
anisotropic materials and perfectly bonded to each other. Ang et al. (2007) developed
axisymmetric model for steady state heat conduction in a multi-material cylindrical
26
system. An analytical solution based on Bessel functions in a series is determined to find
the temperature distribution in the multi-material system. The model developed is utilized
to analyze the thermal behaviors of composites having carbon nanotube arrangement.
The free vibration study of thin FGM cylindrical shells having ring support and
made of stainless steel/nickel alloy was studied by Najafizadeh and Isvandzibaei (2007).
To perform the analysis, ring supports are arbitrarily placed along the shell, which impose
zero deflection in the lateral direction. The TSDT and Love’s shell theory are used to
represent the kinematics and kinetic field and the final governing equations are derived
based on Rayleigh-Ritz method. Type I and type II FG cylindrical shells as considered by
Loy et al. (1999) are considered to execute the numerical part. The frequency behavior of
FG shells indicates a lower fundamental frequency mode for linear value of volume
fraction index. Further, a quite opposite frequency response was visualized between type I
and type II cylindrical shells with respect to circumferential wave number. A remarkable
influence of ring support location on frequency parameter was noticed. When the ring
support was placed at the center, the frequency is maximum for simply supported FG
cylindrical shell, while the frequency tends to show decreasing trend as the ring support
move towards either of its ends.
The free vibration study of FG cylindrical shell with ring supports by Najafizadeh
and Isvandzibaei (2007) was extended to study the effect of various shear deformation
theories on free vibration response of FG cylindrical shells by Najafizadeh and
Isvandzibaei (2009). Numerical studies are performed for cylindrical shells having
different types of boundary conditions (simply supported-simply supported, clamped-
clamped, free-free, clamped-simply supported, clamped-free and free-simply supported
boundary conditions). The TSDT proposed by Reddy was accomplished to establish the
kinematic field and further modified in to FSDT by means of appropriate substitution of
variables in the displacement field. Symmetric pattern of the natural frequency curve is
demonstrated, provided the symmetric conditions are chosen at both the ends along with
the location of ring support at center of the cylindrical shell.
The coupled thermo-elasticity problem of functionally graded cylindrical shells was
solved by Bahuti and Eslami (2007) based on second-order shear deformation shell theory
that considers the influence of transverse shear strain part. The thermal problem was
solved by Laplace technique in time domain, while Galerkin finite element method was
employed for space domain. The heat flux was considered at the inner portion of the shell
to cause maximum temperature. The distribution of axial force for the shell having pure
27
ceramic material was minimum, while it becomes maximum for pure metal plates. For FG
shells, the axial stresses are higher than pure ceramic and metal plates. Further the stress
distribution was found be linear during the shock occurrence, but becomes periodic in
nature after the shock ends and this period follow the pattern of radial displacement.
The buckling of functionally graded circular plates (FGCP) based on HSDT (Reddy
and Khdeir 1989) was carried out by Najafizadeh and Heydari (2008) under uniform radial
compression. They compared their numerical results with the FSDT and CPT results for
different cases and concluded that the HSDT results accurately predict the buckling
behavior of plates, while CPT and FSDT overestimates the buckling loads. They also
observed that mechanical instability of FGM plates are lower than pure ceramic plates at
volume fraction value equal to zero. They concluded that the effect of transverse shear
deformation should be considered as far as thick plates are concerned. Further, the critical
buckling load of FGCP tends to reduce at higher value of volume fraction index parameter.
A C0 higher order formulation has been employed to study the free vibration
analysis of FG curved panels by Pradtumna and Bandyopadhyay (2008). Third order term
of thickness was assumed in the in-plane fields, while constant variation of transverse
displacement (Tarun and Kare 1997) was considered. An element with nine nodal
unknowns was adopted and Sander’s approximation for doubly curved shell was
incorporated in the formulation. Results are presented for cylindrical, spherical and hypar
shells by considering various values for curvature, thickness and volume fraction index.
The frequency declines with respect to lower values of volume fraction index and R/a
ratio. Further dominance of stiffness was observed over mass which results in to rise in
stiffness parameter. In some cases, contribution of both stiffness and mass are observed to
dictate the frequency response of FG panels. For simply supported hypar shells, with the
increase of c/a ratio, abrupt increase of frequency value was observed (c/a=0.5), beyond
this slow increasing trend of frequency was noticed. Further, the superiority of the hypar
shells was established compared to spherical and cylindrical shell while keeping the other
common parameters as constant.
The effects of rotary inertia, normal and transverse shear deformation was
considered by Matsunaga (2008) to analyze the natural frequencies and buckling stresses
of FGM plate. The 2D higher order theory and Hamilton’s principle was used to derive the
governing equilibrium equations. Modal displacements and stresses in the thickness
direction are obtained by satisfying the surface boundary conditions. Integration of three-
dimensional equations of motion has been done to obtain the modal transverse stresses.
28
Modal displacements and stresses are plotted for FG plates considering different thickness
ratios. In addition, magnitudes of internal and external work done for first fundamental
vibration mode are established. Negative sign for internal work was observed due to the
effect of thickness changes in FG plates. Under in-plane stress, lowest displacement mode
gives the critical buckling stress for thin FG plates, while higher displacement modes are
responsible for critical buckling stress in thick FG plates.
The wave propagation technique was employed by Iqbal et al. (2009) to study the
vibration response of circular FGM cylindrical shells. Expressions for strain and curvature
deformations are adapted from Love’s (Love 1952) theory. The magnitude of frequency
was found to be lower for volume fraction index equal to 2 and 3, and for any other
choices of volume fraction index, only a minute variation of frequency was noticed.
Depending on the concentration of stainless steel, nickel and zirconia materials on inner
and outer of shell surface, six categories of FG cylindrical shells are incorporated in the
numerical investigation. It was illustrated that the increment and decrement trend of shell
frequency was dependent on the ratios of Young’s modulus and Poisson’s ratio of the two
constituent materials selected to form an FG shell. But the ratio of density of two materials
does not seem to affect the frequency with regard to value of volume fraction index.
The FSDT in the frame work of element free kp-Ritz method was employed to
study the buckling of functionally graded plates under mechanical and thermal loading
conditions by Zhao et al. (2009a). The exponential variation of effective properties was
considered along the thickness direction. To avoid any shear locking problem encountered
in thin plates, shear and membrane terms are computed using a direct nodal integration
technique, while bending part was evaluated using nodal integration techniques. Different
features such as plate with arbitrary geometry and contain square and circular holes at the
center are investigated. For simply supported and clamped boundary cases, initial rising
trend of critical buckling temperature was observed for volume fraction index (n) equal to
0 to 2, when n rises further and attain the value equal to 5, negligible temperature change
was noticed. Also, for a certain hole dimension and volume fraction index, critical
buckling load for the first mode declines initially, in proportion to hole size and buckling
load exhibit unstable trend as the hole sizes increases.
As an extension of the work by Zhao et al. (2009a), the FSDT combined with
element free kp-Ritz method was performed to study the thermo-mechanical buckling
response of FG shells by Zhao and Liew (2009a). The non-linear through the thickness
distribution of temperature profile was incorporated in the study. Buckling mode shapes
29
are shown for composite panels also having different stacking sequences. The temperature
rise drops for higher value of volume fraction index, and the slope of declination curve
becomes gentle as volume fraction index becomes greater than 2. For the value of volume
fraction index equal to zero, the panel endures a linear temperature field, and buckling
temperature tends to elevate for all the modes as the volume fraction index approaches
higher values.
Zhao et al. (2009c) studied the thermo elastic and vibration analysis of functionally
graded cylindrical shells based on Sander’s FSDT. Variation of axial-stress distribution
was exposed for different boundary conditions of aluminium/zirconia and Ti-6Al-
4V/aluminium oxide plates for various value of volume fraction index. Influence of
constant and modified shear co-efficient on frequency parameter was demonstrated. It was
observed that for R/h=50, the variation between the shear co-efficient was negligible and
for higher values of R/h=100 and 200, the discrepancy is even smaller. The top surface of
the shells ensures tension and bottom surfaces shows compression nature of stresses.
Further, the maximum tensile stress occurs at volume fraction index equals 5, while
minimum value was noticed at volume fraction index equals zero. In addition, mode shape
plots were given for different FGM shells considering various boundary conditions,
thickness ratio and curvature values.
Thermo elastic analysis of FGM plates was performed by Lee et al. (2009) based
on FSDT and element kp-Ritz method. To show the versatility of the method skew and
quadrilaterial plates are also considered. At top maximum compressive stresses are
confirmed for volume fraction index equals 2.0 and at bottom surfaces pure ceramic plate
(n=0) experiences maximum tensile stresses. In addition, difference among various stress
patterns corresponding to different volume fraction index was not significant. The authors
concluded that the effect of length-to-thickness ratio on displacement was independent of
the volume fraction index chosen for the problem. Under thermal loading, negative
deflection was recorded due to the higher thermal expansion of the constituent at the top
surface. When skew angle is large, the higher magnitude of axial stress was induced in the
plate.
Tornabene (2009) presented the FSDT based GDQ method for the free vibration
analysis of conical, cylindrical shell and annular plate structures made of FGMs. Two
different kinds of FGM profiles (FGM1 and FGM2) are proposed based on four-
parameters exist in the power law distribution. With various combinations of these
parameters classical, symmetric and asymmetric volume fraction profile through the
30
thickness can be achieved. Also such profiles lead to the combination of ceramic and metal
at different location of geometry other than the conventional one. Frequencies are tabulated
for first ten modes by varying boundary condition; shell geometry and the parameters exist
in the power law formula. Mode shapes were plotted for the different cases of plate/shell
geometry. It has been seen that for specific values of volume fraction index, FGM1
frequencies are greater than FGM2 model. Because of the curvature effect, increasing the
shell thickness leads to more frequency deviation between FGM1 and FGM2 models. But,
due to lack of principle curvature in the annular plate structures, this effect was not
pronounced. Finally, it has been concluded that the appropriate choice of parameters in the
power law was essential to define the appropriate constituent present at the top and bottom
of the shell/plate structures.
Buckling analysis of thick functionally graded rectangular plates was performed by
Bodaghi and Saidi (2010) using higher-order shear deformation theory. The governing
stability equations are divided into two uncoupled partial differential equations in terms of
boundary layer function and transverse displacement and Levy type solution are employed
to solve the governing equations. When the aspect ratio elevates, the number of half waves
in x direction of critical load gets changed. Since the theory incorporates the shear
deformation effect, considerable influence of thickness parameter was noticed for different
thickness values (h/b). For fixed thickness-side ratio, the buckling load of FG plates lies
between buckling load of ceramic and metal plates, as observed in earlier buckling studies.
The Carrera’s unified formulation for single layer and layer-wise description was
implemented to study the thermo-mechanical response of simply supported FGM shells by
Cinefra et al. (2010). One dimensional Fourier heat conduction equation was solved to
obtain the non-linear distribution profile of the temperature distribution through the
thickness. The principle of virtual displacements was applied to obtain the governing
thermo dynamic equations. The Mori-Tanaka model was employed in the study. Through-
the-thickness variation of transverse displacement component was not constant for thermal
loading, and constant for pure mechanical loading case. Hence for thermal case constant
variation was not valid and this is true for even thin shell cases. In addition the need of
higher order shell theories was exploited to capture all the possible effects of displacement
and stress distributions. In particular, the assumption of higher order thickness was
established under thermal loading conditions.
A three dimensional static solutions are obtained for thick FG plates by Vaghefi et
al. (2010) by assuming the exponential variation of Young’s modulus along the thickness
31
direction. The 3D equilibrium equations are utilized to arrive for local weak symmetric
formulation and the field variables are approximated using the least square (LS)
approximation. In addition, more nodes are incorporated in the thickness direction to
increase the accuracy of the 3D solutions. A wide range of numerical results are presented
by considering different combination of boundary constraints under uniformly and
sinusoidal loading patterns. It was observed that the maximum compressive stress of FG
plates occurs close to the top surface of the plate having low Young’s modulus ratio and
maximum tensile stress of FG plates occurs at bottom of the plate having high Young’s
modulus ratio.
The buckling study performed by Zhao and Liew (2009b) was further extended to
the buckling of conical shell panels by Zhao and Liew (2011) based on FSDT and mesh-
free kernel particle functions. The effective mechanical properties of the FGM conical
panels are assumed to obey power law distribution. Temperature dependent properties of
aluminium/zirconia and stainless steel/silicon nitride panels are considered for the thermal
analysis. When the volume fraction exponent deviates from 0 to 0.5, a fall-off trend was
observed for critical temperature and declines further as the volume fraction exponent
represent the metal segment. When the thickness ratio and semi vertex angle increases,
critical buckling temperature of the panel records declining tendency. This statement was
observed to be common for all the types of boundary conditions. It was stated that the
volume fraction index, boundary conditions, thickness ratio and semi-vertex angle are the
vital parameters that affect the stability of conical panel under thermal loading.
The free vibration analysis of thick functionally graded plates is carried out by
Zhao and Liew (2011). A local Kringing meshless method based on Petrov-Galerkin weak
formulation and combined with Kronecker delta functions was incorporated in the study.
Square, skew and quadrilateral plates are considered in the numerical analysis. The first six
mode shapes are presented for different combinations of volume fraction index and
boundary condition. A pronounced drop in frequency parameter was observed when the
skew angle transform from 30° to 60°, and for other higher values of skew angle, the
frequency drop become insignificant. Further, it was observed that for quadrilateral plates
the in-plane and out-of-plane modes are coupled.
The free vibration behavior of shear deformable functionally graded plates was
studied by Talha and Singh (2011) based on higher order theory and power law variation
of material properties in the thickness direction was assumed. The governing equations are
derived based on variational approach. A C0 element with thirteen degrees of freedom was
32
employed to accomplish the results. To generate the new results, the combination of
various boundary conditions, thickness ratio, aspect ratio, material constituents and volume
fraction index are incorporated. Highest frequency parameter was observed for CCCC
plates, while lowest frequency was discerned for SSSS FGM plates. Also, the frequency
parameter reduces at higher temperature due to weaker Young’s modulus of the material.
The effect of a/h ratio on frequency was more pronounced up to a/h=20, beyond that the
changes are negligible. For a/b greater than 1, the influence of volume fraction index on
frequency becomes insensitive.
Janghorban and Zare (2011) studied the influence of thermal load on free vibration
analysis of Aluminium/alumina FGM plates having different cutouts in their geometry.
Different geometries (square, skew and trapezoidal), different cutout shapes (circular and
rectangular) and various sizes are studied using SOLID 45 and SOLID 70 elements. Under
conventional loading conditions, frequency tends to rise for square/skew/trapezoidal plates,
when the temperature on the upper surface was raised. Further, it was noticed that
increasing the film coefficients of fluid decrease the natural frequencies of skew plate.
The Navier type analytical solution was proposed for static analysis of functionally
graded plates by Mantari et al. (2012b) based on HSDT. Since the theory incorporate the
effect of shear deformation, use of shear correction factor has been eliminated. The
principle of virtual work was employed to derive the governing differential equations.
Uniform and sinusoidal loading conditions are considered to analyze the FGM plates. The
in-plane stress variation for thick and thin plates was observed to be almost same, and
negligible deviation was observed for thick plate case having a/h value equal to 5.
An efficient and simple refined theory that accounts for quadratic variation of the
transverse shear strains across the thickness was proposed by Thai and Choi (2012) to
perform buckling analysis of functionally graded plates. The principle of minimum
potential energy was applied to derive the final governing equations. The variation of
buckling load was observed to be sensitive for lower values of volume fraction index, due
to the higher deposition of ceramic component. Also critical buckling load decreases for
increase in volume fraction index, and increases as metal-ceramic modulus ratio increases.
While plotting the results for influence of aspect ratio (a/b) on buckling load, the variation
is not smooth due to change of critical buckling mode under uni-axial compression, and
becomes smooth under bi-axial compression.
The bending response of functionally graded plates and doubly curved shells was
performed using higher order shear deformation theory and Fourier series based solution
33
methodology by Oktem et al (2012). The variation of mechanical properties of the plate
and shell model was incorporated by means of power law function of volume fraction of
the constituents. The displacement model and strain equations similar to the one proposed
by Reddy and Lie (1985) was assumed. When the plot of transverse displacement vs.
spherical shell curvature was plotted, the decreasing tendency of deflection was observed
as the shell geometry approaches to plate geometry. This effect was due to the
predominance of membrane effects observed in the shell panel. The magnitude of in-plane
normal stress was markedly higher in thin shell panel compared to that of thick shells.
Once again, the membrane effect to reduce the magnitude of stress components was
discerned. The magnitude of transverse shear stress was higher for plates compared to its
spherical counterparts. If the thickness ratio was increases, transverse shear stress deceases
for shells and this is not true for its plate counterpart. Except the value of volume fraction
index equals 0.5, the in-plane stress variation through the thickness was smooth and sharp
trend near the bottom surface was observed in case of both plate and shells. Further, the
effect of curvature shifts the axial stress from the compressive zone to tensile zone
through-the-thickness. It was quite interesting to observe the variation of in-plane shear
stress for moderately deep shell (R/a=10), where for values of volume fraction index
equals 0.5, 1 and 2, the magnitude was maximum at the top of the panel and decreases at
the bottom to approach zero. In addition, in homogeneity also increases the in-plane shear
stress close to the top surface. As a final observation, the predominance effect of curvature
was observed in shells which plays vital role in predicting the static response of the shell
panel.
Tornabene and Viola (2013) obtained the static response of functionally graded
shell and laminated composite shells based on GDQ procedure. Unlike the earlier study
performed by the author (Tornabene 2011), the displacement model has been improved to
consider the geometry of the shell by means of curvature effect in the kinematic as
introduced by Toorani and Lakis (2000). The grading of material properties exist in the
shell layer are defined by a generalized four parameter power law distribution. Two kinds
of power law distributions are shown for the modeling which contains the four variables
that define the material properties of the shell layer at a particular point. GDQ rule was
implemented in the generalized displacement components to estimate the strain and stress
resultants. Further, the 3D elasticity equilibrium equations are solved to get the thickness
profile of the transverse shear and normal stress components. The shell panel consists of
aluminium and zirconia combination of material constituents and six stress components are
34
found for these isotropic materials. Also, two types of power law equations assumed in the
study that generate quite different results in comparison with each other. At the end, the
authors concluded that the higher order terms are necessary in the kinematic model to
catch the realistic static behavior of shell and plate structures.
A higher order theory that accounts for through-the-thickness deformation has been
considered based on radial basis collocation technique by Neves et al. (2013b) for free
vibration response of FGM plates. The principle of virtual work and Carrera’s unified
formulation are combined to arrive for the equations of motion and the boundary
conditions. Results are shown for cylindrical and spherical shells containing simply
supported and clamped boundary conditions. Cubic and quadratic variation of thickness is
considered in in-plane and transverse component of displacement, respectively. Results are
tabulated by considering and without considering the thickness stretching effect. As the
shell geometry transform in to plate geometry, the fundamental frequency reduces for all
the values of volume fraction indexes. The model without incorporating thickness
stretching part, records lower values of frequency compared to the model that include
thickness stretching effect. But for thick plates, the effect of thickness stretching has to be
considered.
The finite element formulation for bending and vibration study of functionally
graded plates was presented by Thai and Choi (2013a), by employing various shear
deformation theories. These theories display strong similarity with the CPT and leads to
four unknowns in the displacement field. The primary variables present in the in-plane and
transverse displacement, respectively, are described by Lagrange and Hermitian
interpolation functions. The contribution of bending and shear component was
incorporated in the transverse displacement field. The in-plane field has been represented
by shape function associated with the derivatives of transverse displacement. Three
different kinds of shape functions described by Shimpi (2002) to include polynomial
functions, Touratier (1991) to include sinusoidal functions, and Soldatos (1992) to include
hyperbolic sine functions are considered in the work. Regardless of boundary conditions
and thickness ratios, the frequency decreases and deflection increases as the volume
fraction index was chosen to represent the bottom of the plate (pure metal).
The buckling problem of thin rectangular FG plates subjected to biaxial
compressive loading with arbitrary edge supports was analyzed by Latifi et al. (2013). The
displacement equations are based on CPT that assumes the form of Fourier double series
(Chung 1981) and physical neutral plane (Zhang and Zhou 2008) was considered to derive
35
the stability equations. The derivatives of Fourier double series was performed by Stroke’s
formulation. It was illustrated that any possible combination of boundary conditions could
be incorporated in the study without imposing any conditions on Fourier series. The FG
plate considered was elastically restrained by means of translational and rotational springs
at the four sides. As expected, they observed that increasing additional constraints on the
boundary increases the buckling load. The results obtained shows that shortening of the FG
plate in the direction parallel to loading direction gives rise to higher value of buckling
load.
A higher order displacement model to include shear and normal deformation effect
was considered to obtain the stress and free vibration response of FG rectangular plates by
Jha et al. (2013a). They considered the material properties in the graded manner along the
thickness direction. Minimization of total potential energy was applied to derive the
governing differential equilibrium equations. They observed the fall-off tend in frequency
parameter for increase in values of aspect ratio (b/a) and pure ceramic plates ensures
maximum frequency parameter. Further, the efficiency of the theory with regard to FSDT
and CPT theory was demonstrated in the study in detail.
The static analysis of functionally graded plates is performed by Castellazzi (2013),
based on nodal integration plate element and FSDT. The power law distribution was
assumed in the study to estimate the mechanical properties of the plate at a specified
height. The interesting outcome from the study was that when the difference between the
material properties of the two material constituent increases, the less clustered stress
profiles are observed about the line of pure ceramic and metal plates. Also, the bottom and
top surfaces of the plate shows more clustered profile of the stress distribution for FG
plates having compressive and tensile nature of stresses at top and bottom of the plate,
respectively.
Tran et al. (2013) proposed isogeometric (IG) formulation for thermal buckling of
functionally graded rectangular and circular plates based on TSDT. To achieve geometric
representation and higher order approximations, non uniform rational B-spline (NURBS)
functions of arbitrary continuous order are employed as basis functions, which also fulfill
the C1 requirement of the HSDT. Due to the stiffness degradation offer by enrichment of
metal part, critical buckling temperature reduces for higher values of volume fraction
index. Under uniform temperature rise, this change was rapid for volume fraction index
nearly equal to 2, and for further values it becomes independent. Also under non uniform
temperature distribution, the FGM plates sustain higher buckling load compared to
36
uniform temperature distribution. Bifurcation type of buckling has been observed in
clamped plates since it neutralizes the bending-stretching coupling. Also, for homogeneous
rectangular plates, non uniform thermal distribution results in linear response of buckling
parameter.
A three dimensional solution was proposed for arbitrarily thick functionally graded
rectangular plates incorporating general boundary conditions by Jin et al. (2014). They
incorporated closed-form auxiliary functions to eliminate all the discontinuities related to
the displacements and its derivatives at the edges of FG plate. Rayleigh-Ritz procedure
was implemented to obtain the exact solution by the energy functions of the FG plate. The
plot of variation of volume fraction through the thickness demonstrates that the volume
fraction varies abruptly near the bottom and top surfaces of the plate for volume fraction
index nearly less than or greater than equal to one. The response of FG plate with several
combinations of realistic boundary conditions was considered in the numerical segment.
A Navier closed form solution based on higher order SSDT developed by Levy and
widely adopted by Touratier (1991) was proposed by Mantari and Soares (2014) to predict
the bending response of functionally graded plates and shells. Stretching effect is
incorporated in the kinematic model and their shear strain shape functions are described by
the two arbitrary parameters m and n which are to be chosen by appropriate displacements
and stress functions. The detailed procedure was explained to extend the theory to non-
polynomial HSDT in FEM. The accuracy of the SSDT is proved in terms of constant
unknowns in the displacement field with reduced error compared to conventional SSDT.
The appropriate values suggested for m and n is 4h (h is the total thickness of the panel)
for some shell cases, while m=n=h/π for some other cases.
A local Kringing meshless method based on Petrov-Galerkin weak formulation
combined with Kronecker delta functions was applied to study the mechanical and thermal
buckling behavior of FG plates by Zhang et al. (2014). To simplify the weak form of
governing equations at the internal boundaries, the cubic spline functions are employed.
Uni axial compression, bi-axial compression, a combination of bi-axial tension and
compression, in-plane shear and thermal loadings are considered as in-plane forces. Under
thermal loading, two types of thermal loads, thermal gradient and heat flux are considered.
It was observed that the buckling temperature attains maximum value for the FGPs
correspond to volume fraction index equal to zero. In addition, higher possibility of
existence of buckling is expected for the case of steeper distribution of temperature in a
FGP and buckling temperature difference between different FGPs increases as the volume
37
fraction parameter decreases. The discrepancy with respect to buckling parameter between
FGP having temperature-dependent and temperature independent properties was noticeable
for higher values of volume fraction index.
The free vibration problem of a two dimensional (2D) functionally graded circular
cylinders was solved by Ebrahimi and Najafizadeh (2014) based on Love’s first
approximation CST. The generalized differential quadrature (GDQ) and generalized
integral quadrature (GIQ) are used to discretize the equations of motion and boundary
conditions, respectively. Two micromechanical models based on Voigt method and Mori-
Tanaka approach are used and the frequency results are tabulated by considering both the
models. It has been observed that the results obtained by both the methods are virtually
same, and shows only a minor deviation with respect to high value of circumferential wave
number. Also, the frequency assessments are prepared for 1D and 2D cylindrical shells,
and it was observed that the 2D FGM shells exhibit improved performance compared to
conventional 1D FGM by means of ensuring high frequency values in all the cases, and
this observation is more obvious for higher value of circumferential wave number.
A brief discussion performed on analysis of FGM reveals the fact that the responses
(static, dynamic and stability) of the FGM plate/shell under thermal and mechanical
loading are greatly influenced by the parameter that dictate the material profile variation
termed as volume fraction index in addition to other parameters like boundary constraints,
geometry, loading condition, thickness ratio, aspect ratio and curvature ratio.
A sinusoidal higher order shear deformation theory is proposed for the bending
analysis of functionally graded shells by Mantari and Soares (2014). The stretching effect
was incorporated in the theory and their strain functions are described by two arbitrary
parameters. A Navier form solution and principle of virtual work are assumed in the
analysis. Extensive numerical results are generated based on sinusoidal HSDT and
compared with the FSDT and other quasy 3D hybrid type HSDT results. Deflection and
stress plots are presented based on the various values of volume fraction index and
geometrical properties of shells.
2.4 STATIC, DYNAMIC AND BUCKLING RESPONSES OF SKEW

PLATES/SHELLS
In general, research studies on skew plates are limited in number owing to the
complexity involved in the analysis. Various solution methods such as trigonometric series
(Echasz 1946; Mirsky 1951), Power series (Raju and Shah 1966; Coull 1967; Iyengar and
38
Srinivasan 1971), polynomial series (Reissner and Stein 1951; Reissner 1952; Stavsky
1963), complex seires (Dorman 1953), biharmonic eigen functions (Morley 1961; Morley
1962), Fourier series (Kennedy and Huggins 1964; Kennedy 1965) have been employed
for analysis of plates having skew geometry, in the past. The technique of finite difference
(Morley 1963; Jenson 1941; Naruoka and Ohmura 1959) has also been successfully used
for the analysis of skew plates, but they have limited accuracy for small skew angles. In
addition, the finite element method found its extensive application in skew plate analysis.
Different studies based on Kirchhoff plate bending elements (Rames et al. 1973; Rossow
1978; Vora and Matlock 1979; Wang et al. 1984; Felippa and Bergan 1987; Ming and
Song 1987) and based on Mindlin elements (Monforton and Michail 1972; Kolar and
Nemec 1973; Ahmed and Mathers 1977; Hughes et al. 1978; Pulmano and Lim 1979;
Hughes and Tezduvar 1981; Belytschko and Tsay 1983; Owen and Figueiras 1983;
Zienkiewicz and Lefebvre 1988; Prathan and Somashekar 1988) have proved better
performance for the analysis of skew plates. In addition to above mentioned techniques,
some other techniques are also available in the literature for the analysis of skew plates.
Among the different methods, variational solution (Morley 1963; Morley 1964; Kennedy
1968; Hadid et al. 1979), electrical analogy (Ruston 1964; Harden and Ruston 1967), point
matching (Warren 1964; Sattinger and Conway 1965), conformal mapping (Aggarwal
1966; Aggarwal 1967), equivalent grid method (Yettram 1972), finite strip method (Brown
and Ghali 1974; Brown and Ghali 1975; Mukhopadhyay 1976; Cheung and Z. Dashan
1987) are few that are reported in the literature.
A Parallelogram-shaped (skew) plates are studied under bending by Butalia et al.
(1990) using a Mindlin nine-node quadrilateral Heterosis element. Uniformly distributed
load, point load and different support conditions are incorporated in the study. In their
earlier research works, the authors proved that the heterosis elements are better than
serendipity and Lagrangian elements with respect to accuracy in case of thin plate
situations (Hughes and Cohen 1978; Hinton and Owen 1984). The moment plots reveal the
fact that, the moment in x and y directions strongly exhibit singularity nature in the vicinity
of obtuse edges having opposite signs. The authors concluded that using H9 (Heteroris
element with 9 degrees of freedom) elements show better convergence for deflection and
principal bending moments at the centre but obtuse corner modeling rather deteriorates.
Hence after comparison, the L4 (Lagraningan element with 4 degrees of freedom) shows
better performance for obtuse corner modeling, even though, only 1/3 the total degrees of
freedom as that of H9 elements are considered for the analysis.
39
Reddy and Palaninathan (1999) employed triangular plate element to perform the
free vibration of laminated skew plates. The consistent mass matrix has been derived in
explicit form and the boundary conditions of the skew corners are implemented through
the transformed element matrices. The fundamental frequency was found to be small for
the layer number of laminate equals 2. Ultimately, if the laminate contains more number of
layers, the response of skew plate under free vibration tends to be that of a homogeneous
orthotropic plate. In skew laminates with simply supported boundary, when number of
layers is greater than 4, the frequency initially rises with ply angle and reaches a maximum
value and decreases again. This maximum value for frequency occurs at ply angle values
45°, 50° and 65° for skew angle values 15°, 30° and 45°, respectively. For all the skew
angle values, the frequency co-efficient elevates with the rise of skew angle due to
reduction of non skew edge distance. The symmetric variation of frequency parameter
found in rectangular and square plates gets distorted for laminates having skew boundary.
Further, the extent of the distortion has direct proportion relation with the skew angle of
the plate.
The FSDT and HSDT (Kant 1982; Kant et al. 1982) based finite element models
are proposed by Babu and Kant (1999) for stability analysis of skew laminated composite
and sandwich panels. For global degrees of freedom of nodes lying on the skew edges of
the plate, the transformation operation has been performed to achieve degrees of freedom
at local domain. The angle ply and cross ply laminates with various lamination parameters,
width-thickness ratios and boundary conditions are assumed to perform the numerical
section. For 90° lamination scheme, the plate with skew angle 45° shows higher buckling
strength compared to 0° lamination scheme. The buckling load factor increases with the
increase in skew angle irrespective of the thickness (for both thin and thick laminates). But
the increase is negligible due to the large transverse shear effect in thick plates. In case of
thin skew laminates, the influence of fiber orientation angle is observed to increase with
the increase in skew angle. But for the case of thick laminates, this observation is more or
less remain same for skew angle 0°,15°, and 30° and reduces for plates with skew angle
value 45°. As far as shear models are concerned, HSDT results are slightly higher than
FSDT results and this is true of skew angle 0°. Also, for SSSS and CCCC laminates, the
effect of shear deformation increases with rise of skew angle and decreases with the
increase of laminate thickness. Regarding sandwich plates, for hf/h≥ 0.05 (hf-thickness of
core and h-total thickness of the plate), HSDT results are the one at lower side to that of
FSDT results and this discrepancy increases with increasing hf/h ratio and skew angle.
40
Hu and Tzeng (2000) performed the stability study of skew composite laminate
plates subjected to uniaxial inplane compressive loads. The finite element based software
ABAQUS has been employed to perform the bifurcation buckling analysis of skew plates.
With the rise of skew angle, critical buckling load of the plate tends to increase. Under the
simply supported boundary, for plates with fibers parallel to edges records lowest buckling
load and quasi-isotropic plates usually show highest buckling loads. When the clamped
boundary was chosen, the buckling results are quite different. Also, for plates with skew
angles, the buckling modes shows more waves in in-plane loading direction. This buckling
mode waves are more for clamped edges than simply supported edges.
A new version of the DQ method was proposed by Wang et al. (2003) for buckling
analysis of isotropic and anisotropic rectangular skew plates. The proposed DQ method
differs from the conventional DQ by means of less degree of freedom used for the corner
points and the direct estimation of weight coefficients. Different features including various
skew angle, aspect ratio (a/b) and boundary conditions are assumed in the numerical
examples. The presented DQ method can also be represented as differential quadrature
element method (DQEM), and useful to solve the problems with discontinuous loads,
geometry and/or mixed boundary conditions.
Haldar (2008) studied the free vibration response of composite skewed cylindrical
shell panels using composite shallow shell element. Effect of shear deformation is taken
into account to avoid the problem of shear locking. The author proposed two types of mass
lumped schemes and one scheme can be applied for free vibration analysis of both thick
and thin shells. Different parameters such as length to radius ratio, skew angles, thickness
to radius ratios, fiber orientation angles and number of layers are varied to perform the free
vibration analysis.
Haldar and Sheikh (2005) investigated the free vibration analysis of isotropic and
fiber reinforced laminated composite folded plates by employing high precision triangular
element. Static condensation technique has been applied to reduce the nodal unknows from
65 to 54. In-plane displacement fields, rotations and transverse displacements are
considered as nodal unknowns by approximating through appropriate polynomials.
Numerical results are presented for composite one-, two-, and four-fold folded plates by
considering different parameters.
A triangular shape element described by three nodes at its corners, three mid-side
nodes on each side and four nodes with the element is employed to study the free vibration
response of plates by Sheikh et al. (2002). Transverse displacement and rotations of the
41
normal have been taken as independent field variables. The technique static condensation
has been applied to reduce the number of nodal unknowns. Authors included the mass of
rotary inertia in a lumped mass matrix. Numerical examples of plates having different
shapes and boundary conditions are solved.
Haldar and Sheikh (2011) employed shear flexible triangular element for the
bending analysis of composite folded plates in the framework of finite element method.
The effect of shear deformation is considered appropriately to eliminate the shear locking
problem. Numbers of results are presented for transverse and longitudinal forces at
different points in addition to results of transverse displacement for various crank angle
values.
Ganapathi et al. (2006) studied the influence of functionally graded materials on
buckling of skew plates in conjunction with FSDT and finite element approach. The
effective properties of the functionally graded materials was assumed to be graded in the
thickness direction and estimated by means of Mori-Tanaka approach and Voigt rule of
mixture. The skew angles ranges from 0° to 45° are considered under in-plane bi-axial
loads for thin and thick plate cases. It was manifested that the Voigt rule of mixture
produce higher buckling loads and decrease in the buckling load is significant up to
volume fraction index ≤2. The volume fraction index beyond 2 yields no significant
reduction in critical buckling load and this tendency is independent of aspect ratio, skew
angle and thickness of the plate. Also, the rate of increase of critical load is high for plate
with a/b=1 compared to a/b value equals 2.
A technical note based on FSDT and finite element approach was presented by
Ganapathi and Prakash (2006) for thermal buckling of functionally graded skew plates.
The temperature in linear and non-linear form was considered across the thickness. The
solution for the thermal problem was obtained by means of one dimensional heat
conduction equation. For thin FGM skew plate, increase in skew angle considerably
increases the buckling strength for a/b=1compared to other higher values of aspect ratio.
When comparison was made between buckling loads under linear and non-linear
temperature variation, non-linear temperature variation yields higher value compared to
linear case. For thick plate case, buckling temperature increases with skew angle and does
not show any appreciable increase with respect to volume fraction index. Further, the
degradation of buckling temperature occurs slowly for the value of volume fraction index
greater than 2.
42
A simple, accurate and reliable algorithm based on discrete singular convolution
(DSC) has been proposed by Civalek (2007) to get the natural frequencies and buckling
loads of composite plates. Four noded element is used to map the straight-sided
quadrilateral domain into a square domain by means of second order transformation and
the equations are finally solved by chain rule. It was observed that the frequency and
buckling load increases as the skew angle of the plate increases.
The vibration study of skew plates using moving least square Ritz method was
proposed by Zhou and Zheng (2008). Due to stress singularities at the obtuse corner of the
plate for large skew angles, the results show slow convergence results. A trial function is
assumed for the transverse displacement field and Ritz method has been employed to solve
for the eigen value equation. The boundary conditions are applied by means of MLS-Ritz
trial function that satisfies the essential boundary conditions along the plate edges. To
overcome the stress singularity problem observed in skew plates more grid points are
placed around the obtuse corners of a skew plate. The authors considered the large skew
angles to perform the vibration study and modal frequencies for different skew angles are
tabulated in the numerical part.
Small amplitude vibration characteristics of thermally stressed laminated composite
skew plates are studied by Singha et al. (2006) using shear deformable finite element. The
possibility of secondary instability is observed in post buckling path. Parametric study has
been performed to study the influences of fiber orientation, skew angle and boundary
condition on the vibration response of thermally stressed composite plates. Due to
singularity observed at the instability points zero natural frequencies are observed.
Das et al. (2004) studied the effects of random road surface roughness on the
impact effects on cable-stayed bridge due to moving vehicles by a zero-mean stationary
Gaussian random process. Finite element method is employed as a planar structure. Each
vehicle in a system is considered as a single degree-of-freedom lumped mass system.
Yadav et al. (1997) performed the dynamic analysis of non uniform section beams with
general and conditions having arbitrary mass and stiffness distributions. Natural
frequencies and modal shape functions are obtained in closed form as an intermediate step.
Kumar et al. (2013) studied the free vibration of skew hypar shells using C0 finite
element formulation based on HSDT. The proposed element has seven nodal unknowns
per node and the effect of cross curvature has been included in the formulation. The
frequency results are generated for different laminations schemes and skew angles. High
frequency values are recorded for skew angle equals 90° for CFCF shells and for other
43
category of boundary conditions, skew angle equals 45° produces maximum frequency.
For hypar shells with combination of free and clamped edges, the skew angle should be
chosen between 45° and 90°, to ensure high frequency.
Jaberzadeh et al. (2013) investigated the buckling of functionally graded skew and
trapezoidal plates under thermal load. The element free Galerkin method was employed
and shape functions are constructed using moving least square approximation. The
technique of orthogonal transformation was utilized to enforce the essential boundary
conditions in the formulation. Different temperature variations such as linear, uniform and
non-linear cases are considered to study the thermal buckling response of FG plates. The
buckling temperature of pure ceramic plates is higher than FGPs, and this effect was more
pronounced in thick plate cases. An interesting observation was discerned regarding skew
plates. As the skew angle rises, the critical buckling temperature increases and the
deviation among skew angle 60° and other cases of skew angle was considerable under
linear, uniform and non-linear cases. Only a small change of critical buckling temperature
was observed for volume fraction index equal to 5 and beyond that no significant
improvement was noticed. This fact is obvious for all the value of skew angles considered
in the problem. Regarding the response of skew angle to buckling temperature, a similar
trend observed in skew plates was concluded for trapezoidal plates also.
Desai et al. (2003) performed the dynamic analysis of multi-layered thick
laminated composite plates using an 18-noded mixed FE model based on layer-wise
theory. Hamilton’s energy principle is employed to obtain the governing differential
equations. In addition to the displacement components, the transverse stress components
have been considered as the nodal degrees-of-freedom to maintain the fundamental elastic
relationship throughout an elastic continuum.
A semi-analytical elasticity solution for accurate estimation of stresses and
displacements in composite and sandwich laminates is presented by Kant et al. (2008). The
assumed FE model satisfies the continuity requirements of displacements and inter-laminar
transverse stresses at the layer interfaces. A two-point boundary value problem (BVP)
governed by a set of linear first-order ordinary differential equations (ODEs) through the
thickness of has been formulated.
Wang et al. (2014) studied the microscopically damaged interface between two
elastic half-spaces using randomly distributed interfacial micro-cracks. It is assumed in the
study that the micro-crack length is a continuous random variable following a probability
distribution. The micromechanical model of the interface solved interms of hyper singular
44
integral equations is used to estimate the stiffness of the interface. Effects of the micro-
crack length and the crack-tip gap between the neighboring micro-cracks are studied in
detail.
2.5 STATIC, DYNAMIC AND BUCKLING RESPONSES OF

SANDWICH PLATES/SHELLS
In general, sandwich plates are constructed by moving the load carrying face sheets
away from the neutral plane or the torsion axis by means of low strength core layer to
increase the moments of inertia of the cross section. Owing to this reason, sandwich plates
registered their application in a variety of engineering field including aircraft, construction
and transportation, where the stiff, strong and light structures are the primary requirements
(Zenkert 1997). Due to the mismatch of material/thermal properties exists at the core-face
sheet layer interface, sandwich plates are susceptible to delamination/debonding type of
failure modes, especially under impact loading (Abrate 1998). In some applications, the
upper layer of the face sheet or core has to be stiffer than the bottom face sheet which
necessitates the implementation of FGM concept for the face sheet and core layers. An
advanced construction of sandwich panel consists of two FG face sheets, not necessarily be
identical, are bonded to a core layer either isotropic/FGM thereby increasing the bending
rigidity of the plate at an expense of small weight. In some cases, the piezo electric effect
has been incorporated in the sandwich construction to serve the purpose of smart materials.
In such situation, piezoelectric ceramics will act as sensors and actuators and usually
placed at the mid layer of the sandwich construction (Shen 2005). In addition, under
thermal environments, the metal-rich face sheets can alleviate the large tensile stresses on
the surface at the early stages of cooling (Noda 1999). In this connection, many research
studies on static, dynamic and stability analysis of FGM sandwich plates are available in
the literature by incorporating graded distribution of material properties either in the core
or face sheet layer.
As an extension of the earlier works submitted on sandwich panels incorporating
functionally graded material under transverse loading (Anderson 2002a; Anderson 2002b),
Anderson (2003) presented an analytical 3D elasticity solution for a sandwich plate with a
functionally graded core. The transverse loading has been applied by means of a rigid
spherical indenter and the contact area and pressure distribution due to indentation was
obtained by using an iterative solution method. The conditions of continuity of traction and
45
displacement components between the layers are utilized to solve the equation based on
Reissner’s theory. The sandwich plate was modeled with orthotropic face sheets and
isotropic core having functional properties that have the exponential variation in the
thickness direction. The plot of in-plane normal compressive stress vs. contact force
establish the fact that no reduction of stress with respect to given contact force was
observed by incorporating stiff material in the core. Also, interfacial transverse shear stress
will not reduce with increase of stiffness ratio Emax/E0 (Emax-maximum stiffness, E0-
minimum stiffness). This trend is common for the other transverse shear stress values. It
was predicted that the increase in the interfacial shear stresses is the effect of localization
caused by the indenter loading to increase the stiffness of the core in the region of core-
face interface.
In part I for deflection and stresses of functionally graded sandwich plates, Zenkour
(2005a) presented the two dimensional solution for simply supported condition. The face
sheets are assumed to have a power law variation of modulus of elasticity and Poisson’s
ratio through the thickness. The core layer is made of homogeneous ceramic material and
by considering the symmetry of the layers, different sandwich plates are proposed. Various
displacement models based on CPT, FSDT, sinusoidal, and TSDT are accomplished in the
study. The pure ceramic plate records small magnitude of displacement compared to pure
metallic plates. The FGM plate undergoes deflection in between that of pure ceramic and
metallic plates by virtue of its intermediate stiffness strength. Under the application of
sinusoidal pressure, the ceramic plate ensures maximum compressive and tensile at the
bottom and top of the plate, respectively. When the plot of transverse shear stress was
considered, the maximum value occurs at a point on the mid-plane of the plot and isotropic
plate shows lower value of stresses compared to FGM plate. For FG plates, the FSDT
theory provides results close to the TSDT and SSDT particularly at the faces of the core
layer.
As an extension of Part I of Zenkour (2005a), Zenkour (2005b) studied the
vibration and buckling response of functionally graded sandwich plate considering rotator
inertia in the formulation. Analytical solution based on sinusoidal shear deformation theory
was incorporated in the study. The sandwich plate modeled herein consists of pure ceramic
material at the core layer and homogeneous face sheets at the top and bottom face sheets.
In the bottom segment, the composition has been varied from a metal-rich part to a
ceramic-rich part, while in the top segment; the composition has been varied from ceramic-
rich part to a metal rich part. The generated results based on SSDT are compared with
46
CPT, FSDT and HSDT based studies. Although the SDPT based frequencies are
marginally lower than elasticity solution; the buckling loads and vibration frequencies
obtained by the SDPT are at considerably higher side than other theories. As the core
thickness with respect to total thickness decreases and volume fraction index increases, the
values of buckling loads and fundamental frequencies are shown to have a fall-off trend.
For the above statement, an exception has been observed for the value of volume fraction
index equals 5. Among the different types of sandwich plates with respect to the
symmetry, the 1-2-1 case exhibit highest sensitivity for the various parameters (aspect
ratio, thickness ratio and volume fraction index) considered in the study. As one may
expect, the uniaxial buckling load may be twice the biaxial one and this observation was
independent of kind of sandwich plate and value of volume fraction index. In general, the
authors concluded that the results corresponding to the ceramic and metal layers are
respectively, the upper and lower bound solutions of those of the sandwich functionally
graded plates.
A higher order based triangular element was employed by Das et al. (2006) to study
the deflection and stress pattern in sandwich plate having homogeneous and FGM layer as
core part. To reduce the computational cost offered by layer wise theories, a single layer
theory has been proposed in which the field variables with weighted average accurately
capture the deformation modes in the thickness direction. To satisfy the inter element
continuity requirement, a hybrid energy functional has been employed and non-uniform
variation of temperature was assumed on the top surface. In homogeneous core sandwich
plates, due to mismatch of properties at the core and face sheets, high magnitude of
stresses are observed at the layer interfaces. Further, steep stress gradients are developed at
the interfaces and to minimize the shear and peeling stresses at the layer interfaces, graded
properties are considered depending upon the temperature distribution dictated by the
extreme environment. In case of sandwich plates with FGM core, due to the occurrence of
high temperature at the top face sheets the plate tend to bulge outwards in all the cases.
The plate corresponding to volume fraction index 0.2 and 0.5 experience a global bending,
where as for the case n equals 1.0 and 2.0, downward expansion of face sheets happens
reflecting the swelling characteristics of the panel. Due to the phenomenon of high values
of thermal strain, the in-plane displacement attains large values at the center of the core.
But top face sheets undergo maximum transverse stresses in case of n=2.0 and this
magnitude was considerably lesser than the one observed in homogenous core model. It
47
was put forward that the discontinuity of strain observed at the interfaces can be effectively
minimized by the reducing the difference in properties of the face sheet and core materials.
Functionally graded rectangular sandwich plates with simply supported and
clamped boundaries are analyzed by Li et al. (2008) based on 3D elasticity solution. Two
types of models viz. the first model with homogeneous core and FGM face sheet and the
later model with homogeneous face sheet and FGM core are considered for the analysis.
The displacement functions are expanded by a series of Chebyshev polynomicals (Cheung
and Zhou 2002; Zhou et al. 2002) multiplied by appropriate functions that satisfy the
essential boundary conditions are assumed. Due to the increase of volume fraction index in
type A or the decrease of volume fraction index in type B simply supported and clamped
plates, the natural frequency decreases with respect to the decrease of material rigidity. The
material rigidity factor plays vital role in thin plates compared to thick plate and further
this effect was little larger for simply supported boundary compared to clamped plates. In
the absence of homogeneous core layer (1-0-1 type plate), the effect of volume fraction
index was more significant than the sandwich plate (1-8-1) having homogeneous hard core.
Also, the role played by volume fraction index is significant where the core was modeled
with hard core instead of soft core. When the displacement plot along the thickness
direction was plotted, flexural and extensional modes are observed and for flexural modes
the displacement is non uniform in nature which implies the existence of normal stress in
the thickness direction. For extensional modes, the deformed plate retains the same
thickness but the in-plane displacement components are symmetrical about the mid-plane.
The stability study of truncated conical shells has been carried out by Sofiyev et al.
(2008) under uniform pressure. The material properties of the three-layered functionally
graded conical shell having FGM core vary in graded fashion through the thickness. This
gradation variation may be arbitrary in nature and combines the volume fraction of ceramic
and metal constituents. Closed form solutions based on Galerkin method are obtained for
the stability analysis of conical shells. The volume fraction of ceramic is often chosen as a
function of linear, quadratic and inverse quadratic term of thickness co-ordinate. As the
ratio of the total thickness to FG layer (h/2a) increases, the dimensionless external pressure
increases for the case of linear and quadratic variation of compositional profile, however
decreases for inverse quadratic compositional profile. But external pressure becomes
insensitive for h/2a greater than 3. On the other side, number of circumferential waves
does not vary with respect to h/2a ratio. When the case of three layer conical shell are
compared with corresponding homogenous conical shell, highest effect was encountered
48
for the quadratic compositional profile (24.46%), while the lowest effect for the inverse
quadratic case (19.15%). The ratio h/2a have constant effect on critical buckling load for
homogeneous case, but have considerable influence on three layer conical shell. For
example, when h/2a ratio equals 1.1, the effect was 24.46% and 19.15% for quadratic and
inverse quadratic change of compositional profile, respectively; further for higher values of
h/2a i.e., h/2a = 6, the effect was 21.44% and 21.18% for quadratic and inverse quadratic
change of compositional profile, respectively.
Part et al. (2008) presented the dynamic response of skew sandwich plate with
laminated composite faces based on HSDT. They have made an attempt to sketch the
influence of skew angle, layup sequence on dynamic response. The authors emphasize the
fact that including higher order terms is necessary to analyze skew laminates, due to the
contributions made by the non-linear shear deformation effects through the laminate
thickness. For skew angle 0° and 15°, the displacement curves are close to each other,
where for the skew angle 30° extremely lower values are noticed. The flexural rigidity of
the plate reduces with the rise in skew angle of the plate. Finally, the authors conclude that
(90°/0°/core), layout may be the best choice while designing cross-ply skew sandwich
laminates.
Brischetto (2009) proposed equivalent single layer and layer wise theories for
sandwich plates with functionally graded core under mechanical loading. The theories are
based on principle of virtual displacement and Reissner’s mixed variational theorem. In the
case of layer wise theories transverse shear/normal stresses are used as primary variables.
For thick plates layer wise models seem to be essential, while equivalent single layer
models are suitable for thin plates with the assumption of higher order expansions. Also,
the use of layer wise models was proven to be good enough to predict the normal stress in
the z direction. If the value of volume fraction index chosen was 10, the use of mixed
models seems to be better choice. Replacement of FGM core exhibit the continuous
distribution of stress components in the z direction, which is otherwise not possible in case
of conventional sandwich plates. The discontinuity offer by FGM in stress case is due to
the application model based on principle of virtual displacement theory, which can be
alleviated efficiently by means of mixed model.
A three-dimensional elasticity solution was presented by Kashtalyan and
Menshykova (2009) for sandwich panels under transverse loading. The core layer has been
modeled with two options; in the first option the core and face sheets are modeled with
homogeneous material having different shear modulus values and the latter option employs
49
functionally graded material in the core part, while the face sheets are assumed to be
homogeneous in nature. The layer with graded properties has exponential variation of
material properties in the thickness direction. From the plot of through-the-thickness
variation of transverse shear stress, the reduction in stress magnitude was observed in the
face sheet/core interface, provided the core portion was modeled by functionally graded
material. But this reduction was observed at the expense of increased transverse shear
stresses in the core part of the sandwich plate. The influence of type of material in the core
is more pronounced in thin panels by means of considerable reduction of in-plane normal
and shear stresses at layer interfaces and face sheets. Also, due to high stiffness offered by
the functionally graded panels the model with such material will considerably reduces the
deflection. The authors emphasize the statement that, the use of functionally graded
material in the sandwich panels, in general, eliminates the deflection and stresses in face
sheets and at layer interfaces.
A three-dimensional elasticity solution presented by Kashtalyan and Menshykova
(2009) for functionally graded sandwich panels was further extended to study the panel
under different loading configurations (Woodward and Kashtalyan 2011). The uniformly
distributed, patch, line, point and hydrostatic loadings are considered in the analysis. The
models similar to the one considered in Kashtalyan and Menshykova (2009) are
incorporated in the analysis. Both the models (core with homogeneous material and core
with functionally graded material), exhibits similar trend as far as variation of out-of-plane
normal stress was considered. This observation is due to the mechanical properties of face
sheets which have major contribution in dictating the stress variation. Further, this
variation has common effect on different types of loading conditions considered in the
study. However, for the panels under point and line loads sharp changing pattern of
stresses are manifested in the core part near to upper face sheets. Further, the magnitude of
transverse shear stresses is reduced if the core with FGM was chosen, and this is true under
all the forms of loading conditions. Under distributed form of loading patterns (hydrostatic,
udl and sinusoidal), the homogenous core shows maximum transverse shear stresses (σ13)
at the center of the core than the one with FGM core. But for point and line loads, the
stresses are maximum at the upper face sheets, and this condition is regardless of the type
of material exist in the core portion. Further, for the case of point and line loadings the
maximum transverse displacement in the homogeneous core increases sharply to attain the
maximum value at the upper portion of the panel, while the assumption of FGM core in the
model diminishes at this peak point.
50
The simple refined theory developed by Shimpi (2002) for isotropic plates and
further extended by Shimpi and Patel (2006a; 2006b) for orthotropic plates was
implemented by Abdelaziz et al. (2011) for functionally graded sandwich plates under
mechanical loading. To derive the governing differential equations, PVD is used and to
obtain the closed form solution of the functionally graded plate with simply supported
condition Navier’s method was employed. Two models the one with FGM core and the
later model with homogenous core and having the different thickness ranges of layer such
as 1-0-1, 2-1-2, 1-1-1, 2-2-1, and 1-2-1are considered. For the case of pure ceramic plates,
different kinds of plates show identical bending behavior. The variation of axial stress is
observed to be very sensitive to the change of volume fraction index. As a general
observation, the pure ceramic plates give smallest shear stresses and deflections and largest
axial stresses. As the value of core thickness with respect to the total thickness of the plate
increases, deflection, axial and shear stresses decreases. Among the different kinds of
sandwich plates, 2-2-1 type FGM ensures smallest magnitude of axial stresses. When the
plot of through-the-thickness distribution of axial stress in x direction was plotted for plate
with FGM face sheets, the stresses are tensile at the top and compressive at the bottom
surface of the plate. This stress variation has linear profile for isotropic plates and non-
linear profile for FGM plates. The plot of shear stress for homogeneous soft core reveals
the maximum value at the mid-plane of the plate and its magnitude is small for FGM plates
than homogeneous metal plate.
Meiche et al. (2011) assumed a sandwich model based on hyperbolic shear
deformation theory for buckling and free vibration study of FGM plates. The final
displacement form proposed by the theory leads to four nodal unknowns in the kinematics
model and the grading technique of the FGM layer follows a simple power law
distribution. Hamilton’s principle and Navier solution are used to derive the governing
equations. The fundamental frequency increases as the core thickness to the total thickness
of the plate decreases and the frequency results are maximum for ceramic plates and
minimum for metal plates. The influence of core thickness has similar impact on buckling
response as that of frequency response. Further, the buckling load increases smoothly in
the plate as the ceramic content in the plate increases.
Merdaci et al. (2011) proposed two refined shear deformation models (RSDT1 and
RSDT2) for the bending response of functionally graded sandwich plates. The theory
developed was variationally consistent and leads to four nodal unknowns in the
formulation. The parabolic variation of transverse shear stresses was assumed and the
51
shear stress conditions are satisfied at the top and bottom of the plate. The core layer was
modeled with pure isotropic (ceramic) material, while the skin layers are assumed to have
graded material properties in the thickness direction. The obtained results for bending
response of functionally graded sandwich plates are compared with the parabolic shear
deformation plate theory, exponential shear deformation plate theory, sinusoidal shear
deformation plate theory and FSDT. Even though all these theories lead to the different
form of displacement field, they provide identical results (either deflection or stresses) for
pure ceramic plates. As the aspect ratio of the plate rises, it tends to elevate the deflection
parameter and this irrespective of the sandwich plate type. The plate having pure ceramic
material ensures highest value for axial stresses in the x-direction and this magnitude
increases with the increase in volume fraction index. On contrary, the ceramic plates have
smallest shear stress values compared to other isotropic and FGM sandwich plates. The
maximum value of shear stresses occurs at the mid-plane and its magnitude for
homogeneous plate (ceramic and metal) is smaller than FG plates. The transverse shear
stress variation of ceramic and metal plates are identical in nature and this is due to the
reason that these plates are fully homogeneous and the stresses do not depend on the
modulus of elasticity of these materials.
After the Part I (Zenkour 2005a) and Part II (Zenkour 2005b) submission with
respect to static, vibration and buckling analyses of functionally graded materials, recently,
Zenkour (2011) presented the solution for thermal buckling of functionally graded plates.
The sandwich plate configuration similar to the one modeled in Zenkour (2005a) and
Zenkour (2005b) was assumed. In addition to the effective mechanical properties, the
thermal expansion was assumed to be graded as per material power law. The non-linear
distribution of temperature profile was considered in the thermal analysis. Buckling results
are presented only for symmetric configuration of sandwich plates. For FGM plates, a
sharp decrement trend of critical buckling temperature was observed with increase in
thickness ratio (a/h) and aspect ratio (b/a) values. But as far as homogenous plate case was
concerned, the decrement was observed in gradual fashion. When the non-linear
temperature variation was opted, the critical buckling temperature produce higher results
compared to uniform variation of temperature. Also, the results corresponding to the linear
temperature change produces the intermediate results between non-linear and uniform
variation cases. When the thickness of the core becomes half the plate thickness, rapid
decreasing trend was observed in buckling temperature to reach minimum values and then
52
increases gradually as per the variation of in homogeneity parameter and this tendency
seem to be exemption for 1-1-1 and 1-2-1 plate cases.
Hadji et al. (2011) presented a four variable refined plate theory (RPT) to get the
frequency response of functionally graded material rectangular sandwich plates. As the
name suggests, the theory contains only four nodal unknowns which was quite less number
compared to other existing shear deformation theories. The sandwich plate having
homogeneous/FGM core with the combination of FGM/homogeneous face sheets are
considered for the free vibration study. Navier’s method and Hamilton’s principle are
utilized to derive the governing equations. It was manifested that the fundamental
frequency of the sandwich plate get decreased with the decrease of material rigidity and
the cause for the observation is the increase and decrease of volume fraction index in type
A and type B plates, respectively. Also, the volume fraction index parameter has vital role
in controlling the frequency parameter in case of thin plates compared to thick plates.
When homogeneous layer of core was considered in the example, the maximum and
minimum results, respectively, are corresponds to ceramic and metal plates. Regarding the
other observations pertains to frequency analysis of functionally graded plates, the results
similar to those of Li et al. (2008) are observed.
Alipour and Shariyat (2012) presented bending and stress analysis of the circular
functionally graded sandwich plates having specific material properties and edge
conditions. The governing equations are derived based on the elasticity-equilibrium
equations in the framework of zig zag theory and each layer of the sandwich plate is
assumed to be made of functionally graded material. The continuity conditions of the
transverse stresses at the layer interfaces are taken to predict the global and local response
of the sandwich plates and also the local variations of the displacements are considered.
The so formed governing equations are then solved by a Maclaurin-type power series
solution. The plot of radial displacement distribution indicates that the plate undergoes
global clockwise bending in addition to counterclockwise local rotation occurs in the core.
Since the ratio of shear to bending deflections was considerable in thick plates, the global
rotation angle will be remarkably affected by the shear forces. Further, as the thickness of
the core increases, the resulting stresses reduces and this reduction was observed to be
more in plate with stiffer core. Similarly, rise in modulus of elasticity of the core leads to
increased stresses in the core and leads to parabolic distribution of the stresses.
Neves et al. (2012c) studied the bending and free vibration response of an isotropic
and sandwich functionally graded plate considering the through-the-thickness variation of
53
deformations. The in-plane field represents the hyperbolic sine terms of transverse
component and transverse displacement contain the quadratic term of some unknown
functions. Carrera’s unified formulation (Carrera 1996; Carrera 2001) has been utilized
and the interpolation operation was performed based on radial basis collocation technique.
An extensive plots and tables are presented for isotropic and sandwich plates (FGM as
core) where the material properties have polynomial material law variation (Zenkour
2006). The results incorporated in the study highlight the importance of thickness
stretching effect to be considered in the kinematic model to accurately predict the
displacement as well as normal stress component.
The bending and free flexural vibration of sandwich functionally graded material
incorporating FGM as core/face sheet has been investigated by Natarajan and Manickam
(2012) by employing QUAD-8 shear flexible element under mechanical and thermal
environment. The assumed kinematic field incorporates the cubic and quadratic terms in
the in-plane and transverse displacement components, respectively, in addition to the zig-
zag function in the in-plane fields (Ali et al. 1999; Ganapathi and Makhecha 2001;
Makhecha et al. 2001). The zig-zag function is piecewise linear at the interfaces and
address the slope discontinuities of u and v at the sandwich interface. The effect of rotary
inertia and in-plane terms are considered for the vibration response of sandwich plates.
Results are presented based on four displacement models (three HSDT models with 13, 11
and 9 nodal unknowns and one FSDT model). The stresses and displacements reduces with
increase of thickness of core layer and increases with rise of volume fraction index. The
increase in ceramic and metal component is the attributed reason for this change in flexural
stiffness of the plate. The first two HSDT models produce identical results for
displacements and stresses, while the latter models (HSDT 9 and FSDT model) cannot
predict the displacements and stress, accurately. For the response of plate under
mechanical loading higher order and lower order models yield identical results for stresses
and displacements. Due to the variation of thermal expansion co-efficient, the models
show different stress variation under mechanical loading. Also, the fundamental frequency
parameter decrease with decreasing gradient index for type B plates, while the fundamental
frequency parameter decrease with increasing gradient index for type A plates due to the
material rigidity difference of the layers. In type A sandwich plates, the material rigidity
decreases with volume fraction index and for type B sandwich plates due to the larger
volume fraction of ceramic material rigidity tends to boost up.
54
The study performed by Neves et al. (2012a) has been expanded to study the static,
and free vibration response of functionally graded sandwich plates. The Carrera’s unified
formulation, PVD and higher-order shear deformation theory that accounts for the
extensibility in the thickness direction was incorporated in the work. Different kinds of
sandwich plates similar to other earlier works have been considered with FGM as core and
face sheet layers. As the thickness ratio increases, the in-plane axial stress increases and
this change was abrupt for the a/h value beyond 10. Also, displacement decreases as the
thickness ratio increases for the different types of sandwich plates and the influence of
volume fraction index is to elevate the deflection parameter. The first ten modes of natural
frequencies are shown for sandwich plates considering thickness stretching effect in the
model. Regarding the buckling analysis observation, fully ceramic plate has higher
buckling strength and this strength decreases as the volume fraction index increases. In
addition, if the core to total thickness of the plate increases the buckling strength of the
sandwich plate increases. Finally, the inclusion of thickness stretching effect in predicting
the static, free vibration and buckling response was established in the study through
various numerical examples and stress variation plots.
Neves et al. (2012a) included the Murakami’s Zig-Zag term (Murakami 1986) to
address the slope discontinuities exist in the functionally graded sandwich plates under
bending by incorporating hyperbolic sine term for the in-plane fields and quadratic
variation term in the transverse displacement field. The Carrera’s unified formulation
combined with the radial basis functions was adopted. The graded properties are
considered either in the core layer or in the face sheets. As the volume fraction index
increases the displacement component increases for simply supported sandwich plate. But
the displacement results again depends on considering or neglecting the warping effect in
the thickness direction. The significant contribution of the warping effect was visualized in
the thick plates, as expected. The transverse displacement has significant rise as the core
to total thickness ratio increases.
An improved higher order theory was implemented by Khalili and Mohammadi
(2012) for the free vibration analysis of sandwich pales consists of functionally graded face
sheets under thermal environment. The temperature dependent material properties are
assumed for the face sheet and core materials by a third-order non-linear function of
temperature (Reddy 1998) and further the distribution of volume fraction was estimated by
power law equation. Unsymmetric and symmetric sandwich plates are analyzed by
Hamilton’s principle. The core in-plane stresses are considered in the vibration response of
55
the sandwich structures. It was manifested that the fundamental frequency elevates with
the increase in the thickness of the face sheets. This observation is due to large amount of
ceramic material and thus increases in the structural stiffness of face sheets. The magnitude
of fundamental frequency exhibit rising trend for lower value of temperatures and for
higher value of volume fraction index. If the non-linear strains of the face sheets are not
considered in the analysis, the influence of temperature on the frequency parameter tends
to fall-off with higher face sheet thickness. This phenomenon is due to the fact that the
thick FG face sheets have more amount of silicon nitride (ceramic) than the thin face
sheets. Also, the temperature has plays more dominant role over pure metal component
(stainless steel), than over ceramic component (silicon nitride). When non-linear terms of
the face sheets are considered in the formulation, the frequency has a tendency to increase
with increase of face sheet thickness. The final conclusion from the study was that the
fundamental frequency increases for higher value of volume fraction index of soft core and
declines for higher value of volume fraction index of hard core sandwich plates.
The bending study of functionally graded sandwich plates is performed by Houari
(2013) based on higher order shear and normal deformation plate theory. The theory
considers the sinusoidal variation of displacements through the thickness and satisfies the
stress free boundary conditions at the top and bottom surface of the plate and thus the
elimination of any shear correction factor. The core is made of isotropic ceramic material,
while the face sheets are made of two-constituent phase of functionally graded material
that obeys simple power law equation. Each displacement (u, v and w) contains the term
for bending, shear and stretching, in which bending and shear terms are functions of x and
y; while stretching part was a function of x, y and z directions. By considering the
symmetry of the plate, three types of sandwich plates with notation 1-0-1, 1-1-1 and 1-2-1
are incorporated in the numerical segment. The influence of shear deformation theories
becomes least significant for fully ceramic plates under the condition of neglecting
stretching effect. When the stretching effect was incorporated in the formulation, the plate
become stiff and hence reduces the deflection of sandwich plates. The axial stress values
are found to be lower in plates that consider the thickness effect than the plates neglecting
the effect, and axial stress increases with the raise in the value of volume fraction index.
A refined trigonometric shear deformation theory that involves four nodal
unknowns was proposed by Tounsi et al. (2013) for bending analysis of functionally
graded sandwich plates under mechanical and thermal loading conditions. The parabolic
variation of transverse shear stress was assumed in the study and satisfies the stress
56
boundary conditions at the top and bottom of the plate. The sandwich plate having FGM
core and homogeneous skin layers was incorporated in the study. By considering the
symmetry of the layers with respect to mid-plane different thickness schemes are proposed
in the numerical part. For all kinds of sandwich plates, deflection declines as the aspect
ratio rises. The difference between the results based on different shear deformation theories
show stable tendency for isotropic case and this fact is irrespective of the types of
sandwich plate. A large variation of response was concluded among different types of
sandwich plates when the case of thermal loading was considered. The theory based on
trigonometric functions produce identical results with sinusoidal shear deformation theory
while almost identical to those of parabolic shear deformation theory. For all types of
sandwich plates, the pure isotropic case produce smallest values of stresses and deflection,
when graded properties are encountered in the plate layer, all the quantities shows rising
trend and this trend depends on the value of volume fraction index. Under various types of
sandwich plates, the plate with symmetric thickness ratio (2-1-2) ensures smallest
transverse shear stresses and the plate with non-symmetric thickness ratio (2-2-1) endures
smallest value of axial stresses.
A meshless based collocation technique has been employed by Xiang et al. (2013)
for the free vibration of sandwich plate made of functionally graded face sheet and
homogeneous core. The proposed meshless method enables to approximate the governing
equations in the plate domain using all the nodes. The modified form of Reddy third order
theory known as nth-order theory has been developed for the purpose. The sandwich plates
of 1-1-1, 2-1-2 and 1-8-1 thickness schemes are considered to generate the numerical
results. Different combinations of volume fraction index and boundary conditions are
considered to tabulate the frequency values.
A 2-D Ritz models are proposed by Dozio (2013) for the free vibration response of
functionally graded sandwich plates having functionally graded core. The formulation
becomes general due to the admissible functions of Ritz variables and the assumption of
invariant properties with respect to the kinematic theory. To generate the results the
boundary conditions other than simply supported and clamped are considered. It was
noticed that irrespective of the different parameters considered in the study, the frequency
parameter generally deceases with the increase in the value of volume fraction index. This
is due to the smaller volume fraction of ceramic component to reduce the stiffness of the
plate. When thin sandwich plates are considered, this effect becomes smooth. When the
value of volume fraction index becomes greater than 5, the first modes for SCSC and
57
CFFF sandwich plates are unaffected. The frequency was found to be higher for the SCSC
plates than the corresponding cantilever plates. This fact is owing to the reason of high
number of constraints imposed in the SCSC sandwich plates. The authors concluded that
the eigen frequencies tend to crowd together with increasing mode number for CFFF
plates.
Yasin and Kapuria (2013) employed four-node quadrilaterial element for static and
free vibration response of multi layered composite and sandwich shells based on efficient
layer wise zig zag theory. The requirement of C1 continuity has been circumvented by
means of improved discrete Kirchhoff technique. Comprehensive examples are performed
by considering various parameters such as boundary conditions, curvature ratio, aspect
ratio and ply angle values. The theory proposed by authors was proved to be more accurate
than three-layer higher order layer wise theories. For moderately thick sandwich shells,
high level of error of the order (> 60%) is observed for the fundamental natural
frequencies, when equivalent single layer theories are incorporated.
2.6GEOMETRICALLY NON-LINEAR RESPONSES OF LAMINATED

AND FUNCTIONALLY GRADED PLATES/SHELLS
In general, the non-linear problem that involves deformations of the order of the
thickness of the plate/shell is called large deformation problem. These problems are
addressed by the assumption of non-linear strain-displacement relations, since the
deformation in the elastic body can have a magnitude that does not overstrain the material.
Because this process was deformation dependent, it was classified as geometric non-linear
problems. In most of the literature works, strain-displacement field was considered by
means of Green-Lagrange strain relation. Further, von Kármán assumptions are imposed
on the strain field by retaining the quadratic terms in the slopes of the deflection and
neglecting other non-linear terms (Reddy 1997), thus leading to final non-linear strain
equations. From the past literatures, even for the case of laminated (Baskar et al. 1993; Vu-
Quoc and Tan 2003; Balah and Al-Ghamedy 2002), homogeneous and isotropic shells
unpredictable response was observed under large deformation situation. Hence, it becomes
vital to study the non-linear response of inhomogeneous materials like FGM plates/shells.
Srinivasan and Bobby (1976) performed the non-linear analysis of skew plates
using finite element method. For the analysis, a high precision confirming triangular plate
bending element was used. The assumed triangular element was initially reported by
Cowper et al. (1970) and extended for non-linear plate analysis by Hwang et al. (1972).
58
The skew angle of the plate ranges from 0°, 30° and 45° are analyzed under large
deformation and it was estimated that the computational effort increases as the skew angle
increases and also less number of elements are found to be sufficient for the clamped plates
than simply supported plate. Except for the case of skew angle 45°, four elements are
required for the analysis of quarter plate. It was manifested that as the skew angle increases
the central deflection decreases for both simply supported and clamped boundary
conditions. This is the observation similar to the one derived in the case of linear analysis.
Under the application of higher loads, the bending stress increases as the skew angle
increases and this trend more marked for the simply supported case. In addition, the minor
principal stress found to be decreased as the skew angle increased and this behavior is
common for both the boundary conditions. But the membrane stresses at the centre are less
for simply supported boundary compared to the clamped one. Since the simply supported
skew plate transfer the load by bending action and hence the reduction of membrane
stresses was observed.
Pica et al. (1980) performed the geometric non-linear analysis of plates using
Mindlin’s theory and finite element formulation. The solution algorithm for the obtained
non-linear equations was based on Newton-Raphson method that combines the series of
linear solutions. Various numerical examples are performed considering linear,
Serendipity, Lagrangian and Heterosis element for square, skew, circular and elliptical
geometry of plates under distributed and point loading. An irregularity sense of
performance was observed in case of quadratic Heterosis (QH) element by means of
providing better stresses at the edge than at the centre. For rectangular type of mesh, the
QH elements are emerged as best among all the other element cases. But the curved
boundaries present in the mesh lead to the inconsistent behavior of all the elements,
particularly in case of predicting the stress values.
Kant and Kommineni (1992) employed the higher order shear deformation theory
for the linear and non-linear finite element analysis of fibre reinforced composite and
sandwich laminates. The transverse shear stresses are assumed to have parabolic
distribution and Green’s strains are considered in the von Kármán sense to account for
large deformations, small strains and moderate rotations. For the finite element analysis, a
simple nine noded Lagrangian quadrilateral element with nine degrees of freedom was
considered. The displacement field considered in the study of Kant and Pandya (1988) and
Reddy (1982) was assumed in the work. The cubic variation of thickness in the in-plane
fields and constant variation of transverse displacement was considered in the kinematic
59
mode. Results are generated by considered various material and geometric parameters of
sandwich laminates. The close range exists between HSDT and FSDT results for thin
cases, and significant deviation has been observed for the case of thick and moderately
thick plates.
Rao et al. (1993) obtained the finite element formulation for the large deflection
analysis of stiffened plates using the eight noded isoparametric quadratic stiffened plate
bending element. To derive the fundamental equations of the plate, the Mindlin’s
hypothesis was considered. The deflection equations under large deformations are based
on von Kármán theory. The obtained non-linear equilibrium equations were based on the
Newton-Raphson iteration technique. The formulation was made for general case, so that
the stiffener can accommodate anywhere in the plate other than nodal line. Further, the
stiffener properties are taken at the Gauss points in the tangential direction of the stiffener.
This leads to the different local axes system which then converted to global axes. The
effect of stiffener in non-linear part was neglected, since it will not have much variation at
the global level. The proposed non-linear formulation based on Mindlin’s hypothesis was
incorporated in the computer code FORTRAN 77 to generate the results. Different
problems such as clamped skew stiffened plate, clamped DRES (Defense Research
Establishment, Suffield) panel, clamped rectangular plate with single stiffener and square
clamped plate are considered under large deformation.
The non-linear transient thermo elastic response of functionally graded plates was
studied by Praveen and Reddy (1998) accounting for transverse shear strains, moderate
rotations and von Kármán strain assumptions. As a general observation it was manifested
that the response of FGM plates are not intermediate to the response of pure ceramic and
metal plates. In the investigation, the shear deformable element developed by Reddy (1984
b) was implemented for the von Kármán strain equations. By imposing the constant
surface temperatures at the ceramic and metal rich surfaces, thermal analysis was
performed and the temperature variation was assumed to vary in the thickness direction.
Two combinations of ceramic-metal constituents were considered in the study so that they
have wide variation of thermal conductivity ratio between the two materials. Thus even the
same values of temperatures are prescribed on the top and bottom surfaces, the temperature
variation for the two chosen cases differ in appreciable manner. At top, the temperature of
300°C and 20°C at the bottom was applied in addition to mechanical loading at the top.
The temperature at any location of the plate for alumina-zirconia plates was lesser than the
aluminium-alumina plates. The thermal distribution was linear for pure isotropic plates and
60
non-linear trend was observed for FGM plates. This non-linear trend will reach its
maximum in terms of the average behavior and again turn back to linear behavior under
some value of volume fraction index. When the intensity of mechanical load increases, the
non dimensional deflection of the plates tends towards negative side of the plate, but for
thermal case the deflection was positive. Due to higher thermal expansion at the top
surface results in the upward deflection of the plate. The various observations concluded
from the numerical example insist the fact that the deflection parameters depends on the
product of the thermal expansion and the imposed temperature. This is the reason why the
FGM plate does not have intermediate response between pure isotropic plates. Since
various FGM plates have close temperature profile among them, the deflection response
was also close to each other. Under the application of mechanical loading, the axial
stresses are compressive at top and tensile at the bottom surface. For different values of
volume fraction index, the FGM plate having n equal 2.0 (represents high content of
ceramic) ensures the maximum compressive stresses at the top.
Sheikh and Mukhopadhyay (2000) obtained the geometric non-linear analysis of
stiffened plates using spline finite strip method. The finite element formulation was based
on Lagrangian coordinate system and the nonlinear equations are formed on the basis of
von Kármán’s plate theory. The final non-linear equations are solved by the Newton
Raphson method and the whole plate was mapped into square domain. Then the mapped
domain was discretized into a finite number of strips where the spline functions are used in
the longitudinal direction and finite element shape functions are utilized in the other
direction. The orientation of stiffener in the plate and the eccentricity are incorporated in
the formulation, so that it can be accommodated anywhere in the plate geometry. The
results obtained are lower than the finite element results, due to the fact of neglecting the
effect of shear deformation in the finite strip method. A wide variety of problems such as a
square plate, circular plate, an annular sector plate, rectangular orthotropic plate, a two bay
rectangular stiffened plate, a five bay Defense Research Establishment, Suffield, Canada
DRES stiffened panel (Houlston and Slater 1986), and stiffened skew plate are treated
under this topic.
An analytical solution for FGM plates and shallow shells was provided by Woo and
Meguid (2001) under thermo-mechanical environment. The solution for the equations was
obtained by means of Fourier series von Kármán assumptions are implemented in the
strain field. A combination of aluminium and alumina was adopted and aluminium plates
are the one that undergoes large deflection. As discussed in the earlier paragraphs, the
61
aluminium plates undergo larger deflection due to the lower modulus of elasticity. But for
n=2.0, even though the plate represents lower proportion of alumina having high stiffness,
it shows better performance in terms of stiffness compared to aluminium plates. Also, the
stress distribution was linear for the case of pure aluminium and alumina plates, while
exhibit non-linear response for other FGM cases. The snap-thorough responses of
rectangular shallow shallow shells are visualized under uniform transverse loading. The
shells with initial less curvature exhibit high stiffness and this stiffness diminishes further
for higher value of curvature, and when the slope of load-deflection curve approaches zero
value, the shell ultimately undergoes buckling. Due to higher thermal expansion, the plate
deflects in the negative direction when temperature field alone was considered. When the
coupling effects are considered in the non-linear analysis, the compressive stresses are
increased about 7.49% at the top of the panel. Wu et al. (2006) obtained the explicit
solution for the non-linear static and dynamic responses of the functionally graded
rectangular plates based on FSDT and von Kármán non-linear assumptions. For the case of
tempeoral discretization Houbolt time marching scheme and finite double Chebyshey
series for spatial discretization are employed. The highest displacement was observed for
aluminium plate and the lowest for alumina plate, because the alumina plate has higher
modulus of elasticity than aluminium plate. As the value of volume fraction index
increases, the value of displacement increases and hence indicating the stiffness
degradation. When pure ceramic plates or FGM plates are considered with volume fraction
index equals 5, the plate with all the edges clamped records lower displacement parameter.
When CCSS and CSCS boundary conditions are considered, the deflection of the CSCS
plate was higher under the value of n=0.0 (pure ceramic case). But when n approaches the
value 5, the displacement for both the boundary conditions was almost identical. The
center displacement rises with the volume fraction index regardless of the boundary
constrains imposed on the plate boundaries. Also, the maximum amplitude of vibration
was higher in case of linear response compared to non-linear responses for all the FGM
plates and the difference in response increases as the n value increases. Similarly, the
deviation in the maximum amplitude of motion between linear and non-linear responses
elevates with rise in n value and this difference becomes highest for metal plates and
lowest for ceramic plates. As a final observation, the volume fraction index equals 2.0, has
more significant effect on the displacement parameter and this observation can be used as a
guiding factor in the design of FGM plates.
62
Arciniega and Reddy (2007a) presented the non-linear geometric problem of
functionally graded shells that consists of two constituents ceramic and metal graded
through the thickness direction. A finite element method with tensor based formulation
having curvilinear coordinates and FSDT are used to model the FGM shell panel. To avoid
the problems of shear, membrane, and thickness locking higher order interpolation
functions are implemented in the formulation. In the Lagrangian formulation, the second
Piola-Kirchhoff stress tensor was used and it conjugates to the rate of Green strain tensor
(Reddy 2004). Different types of problems viz., rollup of functionally graded plate strip,
annular FGM plate under end shear force, pull-out of a functionally graded cylindrical
shell and FGM cylinder under internal pressure are solved in the analysis. By taking the
symmetry of the shell only an octant of the shell has been analyzed in the computational
domain under non-linear response. As a general observation, the shell corresponding to
lower values of volume fraction index (more ceramic) shows higher response than those of
lower values of volume fraction index (more metal). Also the Newton-Raphson scheme
converges below some load level and beyond that it diverges. In general, FGM shells
exhibit identical behavior to that of isotropic and homogeneous counterparts. The bending
response of FGM shells was found to have intermediate response between pure ceramic
and metal panels.
Kordkheili and Naghdabadi (2007) employed updated Lagrangian approach to
obtain the non-linear thermo elastic solution for functionally graded material plates and
shells. The 2nd Piola-Kirchhoff stress was formulated as second-order functions in terms of
a through-the thickness parameter. Under the application of central point load, the pure
aluminium shell ensures greatest amount of deflection, while for FG shell containing
n=0.5, the shell undergoes much less deflection under the same load.
Yang et al. (2008) investigated the non-linear local bending of sandwich plates
modeled as two composite laminated face sheets and graded code under the application of
patch load. The graded core layer has power law variation of material properties along the
thickness direction. The von Kármán non-linearity was incorporated and the kinematic
model was based on the assumption of FSDT. The interaction between the loaded face
sheet and graded core was modeled as an elastic plate resting on a Vlasov-type elastic
foundation. The non-linear bending response was adopted by perturbation technique and
Galerkin method. The final observation from the study was that the use of FGM as core
part will considerably reduce the deformation and local interfacial shear stresses. This was
due to the higher equivalent supporting stiffnesses of the graded core compared to the
63
corresponding homogeneous core. The interfacial shear stresses (σyz) and the deflection are
maximum at the plate center and tend to reduce towards the plate edge. The sandwich plate
with graded core having n=20.0 has the lowest deflection at the center. Always, the linear
solutions over predict the interfacial normal stress and the deflection parameter but
considerably underestimate the interfacial shear stress. This trend of estimation was found
to be more for the case of higher value of volume fraction index. The non-linear local
response of the SSSS and SCSC sandwich plates are almost similar, while lowest
deflection was discerned for CCCC boundary. Both the deflection and interfacial shear
stresses are get affected by the load location and boundary conditions, particularly at their
peak values. As the load moves towards either of the supports, deformed zone and peak of
interfacial stress distributions shift toward the support.
The mesh free kp-Ritz method has been employed to study the non-linear response
of functionally graded ceramic-metal plates under mechanical and thermal loads by Zhao
and Liew (2009b). The von Kármán strains are incorporated in strain part to account for
the small strains and moderate rotations. To model the displacement components, the
FSDT proposed by Reddy (2004) was assumed. To estimate the graded properties of FG
material in the thickness direction, the power law form of distribution was considered. A
stabilized confirming nodal integration method (Chen et al. 2001) was employed to
evaluate the plate bending stiffness instead of Gauss integration. Such an integration will
considerably increase the computational efficiency and also eliminates the problem of
shear locking occurs in case of thin plates. The modified Newton-Raphson method
combined with the arc-length method (Crisfield 2000) was used to trace the complete load-
deflection equilibrium path. As the magnitude of load ranges from 0 to 7 N/m2, the central
deflection of the plates increases with different values of volume fraction exponents. The
domination of stiffening effect may be cited as the reason for this response. But under this
loading magnitude, the response was not linear. Further, as the load range increased from 0
to 20 N/m2, a pronounced non-linear response was observed, whilst maintaining the similar
response of volume fraction index. A close observation on the axial stress distribution
reveals the fact that the top surface of the plate experiences compressive nature of stress
and at the bottom surface a tensile nature of stresses. As a second observation when the
value of volume fraction index equals 2, the plate experiences a maximum compressive
stress. On the bottom portion of the plate, the maximum tensile stress occurs in isotropic
plates, while minimum value observed for FGM plate with n=0.2. When the maximum
temperature of 400°C was imposed on the top of the plate considering different values of
64
volume fraction index, the FGM plate with n=0, experiences the maximum deflection and
n=1.0 corresponds to minimum deflection. As observed from the earlier research works,
the results from the non-linear analysis are always higher than those of linear analysis
results. When the bottom temperature of 20°C and top temperature of 300°C are prescribed
on the plate, an upward deflection of the plate was visualized due to the higher thermal
expansion of the plate at the top. But as the load increases, the upward tendency of
deflection gradually diminishes and results in downward deflection similar to the one
observed in mechanical loading.
The non-linear study performed by Zhao and Liew (2009b) was further extended to
study the geometric non-linear problem of plates using local Petrov-Galerkin approach
based on the moving Kringing interpolation technique by Zhu et al. (2014). The shape
functions constructed by Kringing interpolation method possesses the property of
Kronecker delta function and hence avoids the use of any other special techniques to
enforce the essential boundary conditions. The tangent stiffness matrix was explicitly
developed by meshless technique and the incremental form of non-linear equations was
obtained by the Taylor series expansion. When the load-deflection graph was drawn, a
pronounced non-linear behavior was discerned as the load reaches a certain level of
magnitude. The load-deformation plots obtained for simply supported and clamped
boundary conditions are identical in nature. But as far magnitude was concerned, to get the
same order of magnitude of deflection, the applied loading of the order of eight to ten
times than that of simply supported boundary was imposed. Since the modulus of elasticity
of metal was less than ceramic, the magnitude of deflection increases with the rise in
volume fraction index. When the volume fraction attain very low values (not zero), abrupt
change of axial stresses are observed at the bottom surface occurs, while it happens for top
surface for higher values of n. Further, when the length-to-thickness ratio varies from 5 to
10, the rapid fall-off tendency of deflection parameter was observed beyond which only
little changes are observed. At the higher values of length-to-thickness ratio, an asymptotic
response can be confirmed. Since the analysis was performed under large deformation
case, the portion of deformation resulting from applied temperature is minor relative to the
mechanical response. However, the stress profiles have considerable effects under thermal
loading. Considering the thermal dependent properties gives rise to more accurate thermal
field and hence the possible reconstruction of stresses distribution within the FGP.
As an elaboration of the earlier study by Zhao and Liew (2009 b), the non-linear
study was performed for FGM shells by Zhao and Liew (2009 c) under thermal and
65
mechanical loading conditions. The Sander’s non-linear shell theory in combination with
von Kármán strain was incorporated in the formulation. The solution methods that are
already discussed in the work of Zhao and Liew (2009a) are incorporated in the work.
Because of the low stiffness of the panel at higher values of volume fraction index the
deflection shows increasing trend. Except for the case of pure ceramic and metal plates,
there was not much difference for the lower limit loads for FGM panels of various volume
fraction indexes. The deflection response of the FGM panels lies in between the responses
of ceramic and metal panels. The maximum compressive and tensile stresses occurs for the
values n=2.0 and n= ∞, respectively. Similarly, the minimum compressive and tensile
stresses are observed for the values n=0 and n=0.2. Also, the drastic stress changes are
observed at the bottom compared to top of the panel. When a temperature of 200°C was
imposed on the top of the panel, the panel experiences the compressive stresses through the
thickness with the exemption of portion near the bottom where the stresses are almost zero.
Again, the response under thermal environment was intermediate to that of isotropic and
homogeneous plates.
Wankhade (2011) presented the geometric non-linear analysis of skew plates which
require more computational effort due to the existence of singularities involved at the
obtuse corner with varying skew angle. The finite element formulation used in the study
considered the transverse shear effect by considering Reissener/Mindlin thick plate theory.
The deflection of the skew plate increases as the skew angle elevates and the load
deflection curve for 60° angle approximates to straight line. Hence increasing the skew
angle increases the rigidity of the plate and hence increases the overall strength of the
structure. The variation of membrane stress was almost straight lines for all the skew
angles and considerable deviation in the membrane stresses are observed for skew angle 0°
and 60° under the same magnitude of loading. But the curve of bending stresses exhibit the
curved tendency for all the skew angles. When the membrane stresses are plotted for
different aspect ratios (b/a), the membrane stresses are not affected by large value of skew
angles.
The non-linear response of FGM plates was reported by Singha et al. (2011) using
a four node high precision plate bending finite element. The FSDT theory with the exact
location of neutral plane was incorporated and shear correction factors are estimated by
using energy equivalence principle. The value of in-plane stresses evaluated from the
constitutive relation and three-dimensional equilibrium equations are used to estimate the
transverse shear and transverse normal stress components. To predict the load-
66
displacement response curve, the Newton-Raphson iteration method was employed. To
avoid shear locking, shear strains are considered as nodal unknowns. When the in-plane
normal stress variation was plotted, the isotropic plates exhibit linear variation, while FGM
plates show non-linear variation due to the variation of Young’s modulus through the
thickness. A hardening type of non-linearity was observed for simply supported FGM
plates, and this non-linearity was less with the increase in the value of n due to the low
stiffness of the plate. Depending on the value of volume fraction index, the maximum
transverse shear stress decreases with the increase in load parameter. The in-plane stresses
are compressive in nature at the top and tensile in nature at the bottom of the plate. With
the increase of transverse displacement, the in-plane stress variation becomes highly non-
linear depending upon the value of volume fraction index. For immovable in-plane
boundary, the degree of hardening non-linearity was more compared to movable in-plane
boundary, as anticipated.
A four variable refined plate theory proposed by ABDELAZIZ (2011) has been
accomplished by Fahsi et al. (2012) to study the non-linear cylindrical bending of
functionally graded plates under thermal and mechanical loadings. The material properties
are assumed to vary in the thickness direction according to simple power law distribution
in terms of volume fraction of material constituents. To account for the effect of geometric
non-linearity von Kármán assumptions are incorporated in the work. The minimization of
total potential energy was implemented to get the final governing equation for non-linear
analysis. Different examples are performed by considering various material and geometric
parameters.
Malekzadeh and Heydarpour (2012) obtained the thermoelastic transient response
of functionally graded cylindrical shells under moving boundary pressure and heat flux.
The temperature dependent material properties are considered in the radial direction. To
incorporate the non-Fourier effect, hyperbolic heat conduction equation was utilized that
include the influence of finite heat wave speed. A combination of GDQ and FEM was
employed to solve the governing system of equations. The resulting non-linear equations
are solved using Newmark’s time marching scheme in temporal domain. It was observed
that the volume fraction index has significant effect over radial displacement, tangential
and axial normal stress components. But the influence of radial stress component was
small and can be ignored, especially in the regions that are far from the inner portion of the
shell. Except on the radial stress component, increase in the length of the FG cylinder has
considerable influence over other parameters. The effect of the length of the cylinder on
67
transient response was much more significant than steady state response. As far as the
influence of thickness parameter has concerned, its response increases with the increased
elapsed time. When the effect of thermo mechanical load front velocity was studied for
clamped FG cylindrical shells, increasing the velocity increases the results (displacement
and stresses). Also, as the time level increases, the-thorough-thickness variation of
clamped FG cylinder results approaches to steady state values.
A finite element solution incorporating shear and normal deformation effect in the
soft core sandwich plate model was presented by Madhukar and Singha (2013) for
geometrically non-linear and vibration response. The von Kármán assumptions are
introduced in the strain equations and in-plane and rotary inertia was considered in the
equations of motion. Newton-Raphson iteration technique and harmonic balance method
are employed, respectively, for static and vibration analysis. The displacement model
proposed by Kant and Swaminathan (2001) and Kant and Owen (1982) was assumed in the
model. The components of shear strain vectors are taken as separate nodal unknowns and
thus avoiding the problem of shear locking. If the thickness of the soft core elevates, the
displacement parameter also tends to rise for the problem under consideration. Due to the
influence of more shear and normal deformation effect, the central displacement increases
with the increase of core-to-face thickness and span-to-thickness ratio, for a fixed load
parameter. Regarding non-linear vibration analysis, the frequency ratio is more for thick
plates compared to the case of thin plates. If the total thickness of the plate has kept
constant, the non-linear frequency ratio gets decreased with the increase of core thickness.
Also, the excitation frequency if increases from zero or decreases from higher value, the
flexural vibration amplitude increases. Since structural damping was not considered in the
study, the non-linear flexural vibration amplitude increases in rapid manner as the
excitation frequency approaches the linear flexural vibration of the plate from either side.
Upadhyay and Shukla (2013a) investigated the nonlinear static and dynamic
response of functionally graded skew plates using HSDT and von Kármán non-linear
kinematics. The chain rule of differentiation and linear mapping was performed to
transform from physical domain into computational domain. The displacement model
proposed by Kant and Pandya (1988) was implemented in the study. It was noticed that the
aluminum plate exhibit less stiffness behavior with respect to the FGM skew plates. The
difference in displacement parameters of the 90° and 60° plates (23.35%) was considerably
lower than the displacement difference between the 90° and 30° (83.07%) skew plates.
When compared to square plate the deflection decreases by 29.1% and 88.45%,
68
respectively for 60° and 3°0 skew plates for the linear variation of volume fraction index.
In addition, for 30° skew plates irrespective of the value of volume fraction index the
displacement decreases. The deflection pattern of the skew plate considering various types
of boundary conditions was similar in the pattern. Under transient response, amplitude and
the time period of motion decreases with the fall-off value of skew angle. But this behavior
was no longer exists for higher value of volume fraction index i.e., with increase in volume
fraction index, time period of motion decreases and amplitude tends to be increased. Due
to the coupling effect of stiffness during increase and decrease of volume fraction index,
the stiffening effect of the plate decreases with the skew angle. When different patterns of
loading are considered, rectangular pulse loading ensures highest motion of amplitude,
while the lowest ensured for exponential pulse, depending upon the loading curve area.
Upon removal of the load, the amplitude corresponding to sine pulse increases, while it
remains same for other loading forms.
Kaci et al. (2013) solved the non-linear bending problem for sigmoid functionally
graded plates in which the variation of material properties is considered in the thickness
direction. The governing equations are reduced to a linear differential equation with
nonlinear boundary conditions. Under the application of pressure loading, the stresses are
found to be compressive at the bottom and tensile at the top surface. Further, high
magnitude of tensile stresses are confirmed on the top surface for linear analysis at n=2.0;
while under non-linear analysis, same location for maximum was observed but for n=0.2.
The stress profiles for pure metal and ceramic plates are always linear irrespective of the
type of analyses. When FGM plates are modeled by sigmoid distribution of material
properties, response of such plates are identical to those of homogeneous plates.
Duc et al. (2014) employed the classical shell theory taking into account the
geometric non-linearity, geometric imperfection and Pasternak type elastic foundation for
the analysis of shallow spherical shells under mechanical and thermal conditions. The
influence of elastic foundation, external pressure, temperature, geometric and material
properties on the non-linear buckling and postbuckling of the shells was presented in
detail. When immovable boundary conditions are considered, the snap-through behavior
the FGM spherical shell becomes more unstable. When the effect of elastic foundation was
incorporated, the snap through curve become more stable. When temperature field was
prescribed on the surface, outward deflection was confirmed, which is the identical
statement confirmed in other related studies. As soon as the interaction of mechanical part
was encountered, out ward deflection tends to reduce and external pressure exceeds
69
bifurcation point of load thus results in an inward deflection. It was concluded that the
ability of the system has been reduced in the presence of temperature field.
A cell based smoothed Mindlin plate element was recently proposed by Van et al.
(2014) for geometric non-linear analysis of functionally graded plates. The C0 based non-
linear formulation was developed and von Kármán strains are implemented in the analysis.
A simple two step procedure was incorporated to analyze the plates under mechanical and
thermal loads. As observed in earlier research works, under the application of thermal load,
an upward deflection of the plate was observed due to the higher thermal expansion of the
plate at the top surface. When the volume fraction index approaches high concentration of
metal segment, an elevated deflection parameter was calculated. Also, when the
temperature at the ceramic surface increases, the deflection of plate becomes small.
Different plots for axial stress variation through the thickness were established for different
material combination and temperature range.
2.7 POST BUCKLING RESPONSE OF LAMINATED AND

FUNCTIONALLY GRADED PLATES/SHELLS
The geometric instability of the structure under in-plane thermal/mechanical load was
identified as buckling. Thin walled structural members are able to carry additional load
after buckling, known as post-buckling strength before failure of the structure. In order to
utilize the full strength of the plate and shell structures under mechanical and thermal
environments, it is necessary for a designer to know the actual critical buckling load
(primary bifurcation) and the corresponding post buckling strength (secondary
bifurcation). When FGM plate/shells are employed as heat-shielding components with
restrains against in-plane expansion and contraction, considerable amount of strains are
induced and stresses are developed at elevated temperatures. This situation establish a
certain membrane pre-stress that may commence the buckling and post buckling
phenomena in the structure. In addition, the asymmetric material properties with
temperature functions make the post buckling response of the structure more complicated.
Hence thorough understanding of the response of the FGM plates and shells in the post
buckling region seems to be vital for optimum and effective design of FGM structures.
70
A finite element based Mindlin shallow shell formulation was presented by Pica
and Wood (1980) for circular and square plates under in-plane loading. In addition to in-
plane axial loads, the square plates are assumed under in-plane shear loading. The
problems of cylindrical, spherical shells and shallow shells are also treated, which records
the snap through behavior. The variation of central moment, membrane stress, and
deflection with respect to applied load are plotted for different shell types and boundary
conditions.
Yang and Shen (2003) obtained a semi-analytical approach for the large deflection
and post buckling response of functionally graded plates under in-plane and transverse
loading conditions. For the analysis, the temperature dependent material properties are
considered that obey the simple power law distribution in terms of the volume fraction of
the constituents. The CPT based model with the effect of plate foundation interaction was
considered where the Winkler elastic foundation can be treated as limiting case. The plate
with two edges clamped and the remaining two edges simply supported clamped or it may
have elastic rotational edge constraints was modelled. The plate with intermediate
properties (FG plate) has intermediate response (deflection) and the ceramic plate exhibit
lowest deflection because of the highest stiffness. When Bending moment plots were
considered, this tendency of plates no longer exists. As the in-plane compressive load ratio
rises, both deflection and bending moments shows elevating trend. A deep insight into the
load deflection and load-bending moment curves insist that the more rigid boundary
constraints ensures deflection and bending moments of small magnitude. The post
buckling response of the FGM plates are identical to that of corresponding non-linear
bending curves. In the post buckling region, the load capacity of the FGM plate increases
with the increase of edge rotational rigidity or with the increase of foundation stiffness.
When the clamped FGM plates are subjected to uni-axial compression, the deflection
deviates abruptly under the effect of initial transverse pressure. However, beyond the post
buckled range the response of initially pressurized plate was almost asymptotic to that of
pressure free plate.
An analytical solution was presented by Woo et al. (2005), for the post buckling
behavior of moderately thick plates and shallow shells under temperature field and edge
compression. The HSDT and von Kármán type non-linearity was assumed in the analysis.
The mixed Fourier series solution was obtained and the results are presented for Reissner-
Mindlin theory and CPT. The higher order theory developed by Reddy (1984b) for
moderately thick plates was assumed in the analysis. It was ascertained that to predict the
71
buckling load in accurate sense higher order theory should be incorporated in the analysis,
in particular, when the thickness of the plate equals the one-twentieth span of the plate.
Also, considering the shear deformation terms in the theory leads to the under estimation
of critical buckling loads. In spite of the higher order terms present in the theory, pure
isotropic plates exhibit linear response, while FGM plates shows non-linear response of
post buckling curves. The critical load value for a cylindrical shell was higher than
corresponding flat plate and when the shell reaches the critical load, the snap-through
buckling happens and shell structure collapses. When bending moments are calculated
under edge compressive loads, for FGM shells, the magnitude was not zero before the
buckling because the clamped edges prevent the transverse deflection from occurring.
Under simply supported boundary condition, the isotropic plates (ceramic and metal
plates) exhibit conventional type of buckling and heterogeneous plates deflect transversely
due to the structural asymmetry of the middle surface. At the same time, the FGM plates
show very high resistance at the early stages of deformation.
Yang et al (2006) performed the thermo-mechanical post buckling analysis of
functionally graded cylindrical panel considering the temperature-dependent properties in
the thickness direction. During the initial stage, the panel was stressed by an axial load and
further subjected to the change of temperature in uniform sense. The non-linearity was
accounted in von-Kármán-Donnell type and the CPT based kinematic model was
incorporated. To trace post buckling equilibrium path, a semi analytical method based on
differential quadrature was utilized with an iteration process. When both (temperature in-
dependent) TID and TD (temperature dependent) properties are assumed for the particular
problem, TID solutions are higher (about 9-18%) than the corresponding solutions
considering TD properties. In connection with this, lowest buckling temperature are
observed for metallic panel (nickel plates) compared to FGM and pure ceramic plates
(silicon nitride). Due to the sharp decline tendency of stiffness at higher n values, buckling
temperature increases for different values of n. The thermal buckling capacity of the
cylindrical panel was enhanced with the presence of axial tensile pre-stress; whereas the
opposite behavior was visualized under the axial compressive loads. As the known fact, the
buckling temperature was smaller when the edges are fully restrained against any in-plane
movements. When the panels are initially stressed under axial compression, the post
buckling paths of the simply supported panels are not bifurcational, and in the presence of
axial core, initial deflections are induced so the post buckling path do not start from the
coordinate origin. Further, the post buckling temperature difference between the TID and
72
TD solutions are higher for CCCC panel than that of the SCSC panes, but the former type
panels have highest post buckling load carrying capacity compared to its SCSC
counterpart.
Wu et al. (2007) obtained the analytical solution for the post buckling response of
functionally graded plates under mechanical and thermal loading by means of fast
converging finite double Chebyshev polynomials. The mathematical model was based on
the FSDT and von-Kármán non-linear kinematics. The critical buckling temperature and
buckling load reduces with the presence of higher content of metal i.e., higher value of
volume fraction index. For all the cases performed in the study, it was manifested that up
to the value of volume fraction index equals 2, a significant impact on buckling and post-
buckling response was visualized. Also, the buckling temperature of FGM plates was
found to have lesser values compared to the pure isotropic plates (ceramic) and this is
irrespective of the boundary conditions. When the buckling and post buckling strength of
the plates vs. volume fraction index are plotted, the deviation between the buckling and
reserve strength of the plate for n=5.0 and 10.0, was very less. The pure ceramic and metal
plates, respectively, ensures higher and lower buckling and post buckling strength with
respect to various ranges of volume fraction index. As the plate aspect ratio tends to
elevates, the buckling responses of the plate decreases. The performance of square plate
and plate with aspect ratio equals 1.5 are almost identical for n=0 and 2. Similarly, the case
of aspect ratio b/a=3 and 4, indicating the beam response of the plate beyond the value of
b/a=3. When different boundary conditions are considered, the FGM plates record lower
buckling and reserve strength compared to alumina plate.
Panda and Singh (2009) employed the Green-Lagrange based HSDT for the post
buckling analysis of laminated composite cylindrical/hyprboloid shell panel subjected to
uniform temperature field. The non-linear stiffness terms exists in the Green-Lagrange
relationship was considered in the formulation. The solution of governing equations was
obtained by minimizing the total potential energy of the system. The quadratic variation of
transverse shear strains and transverse shear stresses are considered in the displacement
model (Reddy 2004). The critical buckling temperature generally decrease with the rise in
R/a for all the laminations, and increases with the increases of layer numbers. When
different values of curvature ratio (R/a) and amplitude ratios are chosen for the problem,
the temperature ratio becomes maximum for all the laminates at R/a=100. Due to the
severity of non-linear effect, the post buckling strength do not show any definite trend over
the amplitude ratios (some places shows increasing trend and at some places decreasing
73
trend). The primary bifurcation strength was highest for the case of anti symmetric angle-
ply lamination scheme for the chosen thickness ratios except for thick panels (a/h=10). As
far as hyperboloid panels are considered, as the aspect ratio increases, the buckling
temperature parameter decreases due to the tendency of the panel to become flat under
higher aspect ratios and curvature ratios. Also, the post buckling strength for square panels
are higher than that of rectangular panels. Except for the case where amplitude ratio 1.2
and R/a=20, the hyperboloid panels ensures higher post buckling strength compared to
cylindrical panels. As different modular ratios are chosen for the hyperboloid panels, the
buckling temperature decreases with increase in modular ratio and further follows mixed
type of trend with different values of amplitude ratio. In general, the temperature ratio
increases with the increase of amplitude ratio and decreases with the increase of modular
ratio for both cylindrical and hyperboloid panel, with few exceptional cases.
The element free kp-Ritz method in conjunction with the FSDT was adopted to
solve the post buckling problem of functionally graded plates under edge compression with
temperature dependent properties by Lee et al. (2010). The displacement fields are
assumed by means of kernel particle functions and direct nodal integration method was
employed to evaluate membrane and shear terms thus to avoid any shear locking problem.
In conjunction with the modified Newton-Raphson method the arc-length iterative
algorithm was considered for the solution of non-linear equations. The highest and lowest
magnitude of load factors are ensured for ceramic and metal plates, while the frequency of
FGM plate with n=0.5, 1.0 and 2.0 fall between those of pure isotropic plates. When the
influence of boundary conditions are considered on post buckling responses, initial curve
shows no significant different, as the load increases, the remarkable deviation between the
curves was discerned. Under thermal analysis, a prescribed temperature of 20°Cand 100°C
was applied on the bottom and top of the shell panel, respectively. When the plate
subjected to temperature field in addition to the temperature field, the plate experiences the
higher post buckling deformations due to the thermal force and moment resultants that
reduces the stiffness of the plate. When different temperatures are imposed on the top of
the plate, the plate with highest temperature undergoes large initial deflection and ensures
a greater amount of post buckling deformation. At high temperature, the stability of the
plate tends to decrease and thus leads to the larger deformation at post buckling stages.
Liew et al. (2012) presented the post buckling analysis of functionally graded
cylindrical shells under thermal loads and axial compression based on FSDT and element
free kp-Ritz method. The strain field based on von Kármán assumptions and power law
74
distribution of mechanical properties in the thickness direction was considered in the
analysis. The other solution techniques similar to that of Lee et al. (2010) are incorporated
in the study. When the plot of central deflection vs. volume fraction index was studied, the
deflection increases slowly at initial load condition for n=0 and shows fast increasing trend
and finally shows monotonically increasing trend for higher load values. The degree of
displacement at higher volume fraction index was large due to the effect of stiffness
degradation. Under the plot of end shorting vs. load curves, after certain point of loading,
the curve exhibit negative slope. From the various numerical examples presented in the
study, it was manifested that the curvature effect of panels plays a vital role in predicting
the post buckling response of FGM shells.
The buckling and post buckling responses of laminated composite plates was
performed by Dash and Singh (2012) using HSDT in conjunction with Green-Lagrange
strain-displacement relationship. All the higher order terms present in the Green-Lagrange
relations are included in the analysis. The load ratio (post buckling load/critical buckling
load) results for a simply supported plate for different stacking sequences indicate that the
load ratio increases with the increase in amplitude ratios. As the two layered square plate
was chosen, the buckling resistance was found to be less up to the value of amplitude ratio
0.8. When load ratio pattern for different amplitude ratios are tabulated for clamped plate,
initially the load ratio increases with the amplitude ratio and beyond that no definite trend
was discerned. This trend was not same for the different kinds of laminated plates chosen
for the problem. Next, the influence of orthotropicity on the load ratio was studied for
cross ply square laminates. The buckling resistance increases with the increase of
orthotropy and amplitude ratio chosen for the plate. Further, no definite trend was observed
regarding the influence of orthotropicity upto the amplitude ratio 1.0. But, it shows
increasing tend beyond the value 1.0.
Considering the randomness in the material properties, Lal et al. (2013) presented
the post buckling response of functionally graded plates under thermo-mechanical
environment. Two variables i.e., material properties of each layer and volume fraction
index are assumed as independent random input parameters. A C0 based HSDT and von
Kármán non-linear kinematics was incorporated in the study. To solve the issue of C1
continuity of the HSDT, the authors assumed the Co continuous element as incorporated in
the work of Singh et al. (2002). Thus the artificial constraints should be enforced
variationally through the approach of penalty method, but as per the study of Shankar and
Iyenger (1996), accurate results for C0 formulation are presented without enforcing any
75
penalty approach. Under the thermal part, the material with temperature independent
properties (TID) and temperature dependent properties (TD) are incorporated. The co-
efficient of variation (COV) with TD material properties are highly sensitive compared to
TID material properties. The post buckling response of FGM plates are more affected by
the random change in COV of different parameters viz. Ec, Em and n. Therefore, the strict
control of these parameters was vital to achieve the reliability of the FGM plate. The FGM
plate with TID material properties shows high value of mean dimensionless post buckling
load, while lower value of mean dimensionless post buckling temperature was confirmed
for TD material properties. For different amplitude ratios, the post buckling load and
temperature increases with the rise in the value of amplitude ratio.
Upadhyay and Shukla (2013b) presented the buckling and post buckling response
of laminated composite and sandwich skew plates based on HSDT and von Kármán
strains. The linear mapping technique was employed to transform the physical domain
into computational domain. The governing equations and boundary conditions are
discretisized in spatial domain using finite double Chebyshev series solution. The
displacement model proposed by Babu and Kant (199), where the in-plane fields have
cubic term and constant for transverse displacement are considered in the formulation.
Under the action of uni-axial compression, as the skew angle of the plate rises, both the
buckling and post buckling strength of the skew plate increases and response of symmetric
cross-ply skew plates are higher than anti-symmetric plates. At low skew angles, the
lamination scheme of the plate has more pronounced effect on post buckling strength. But
no deviation in the responses (buckling and pot buckling strength) was observed for cross-
ply and angle-ply plates, when the plate was subjected to uni-axial compression. Further
when the case of positive in-plane loading was considered, except at skew angle value 45°,
the cross and angle ply laminates shows no deviation in the post buckling response.
Under negative in-plane shear loading, anti-symmetric skew plates exhibit higher
buckling strength that that of symmetric skew plates. Also, the post buckling strength of
the skew plates are higher for the case of positive in-plane shear loading than that under
negative shear loading. This may be due to the development of tensile stresses at the acute
corners of the skew plate under positive in-plane loading. In addition, the influence of
boundary conditions and thickness ratio on post buckling strength was also studied for
laminated skew plates. The sandwich skew plats are treated under uni-axial compression,
bi-axial compression, and positive and negative in-plane shear loads. For the case of uni-
axial compression, no significant improvement has been observed for the skew plates
76
having high core thickness and high skew angle. In conclusion, the sandwich plates even
though exhibit higher buckling strength under in-plane loads, no significant improvement
was observed. Hence to get the higher reserve strength for sandwich skew plates, the ratio
of material properties of core to face sheet should not be too high.
2.8 APPRAISAL FROM PREVIOUS RESEARCH WORKS

A lucid collection of literatures dealing with the responses of FGM structures by
considering different parameters has been presented in the previous sections. A deep
insight in to the available literatures reveals the availability of different studies related to
static, dynamic (free/forced vibration) and stability of plate and shell structures made of
FGM components. In addition, a finite volume of research works can also be located in the
literatures on geometric non-linearity and post buckling problems of FGMs. A very limited
number of studies are presented considering skew geometry and multilayer concept in
FGMs. However, the critical review of literature manifests the need for efficient and
appropriate model to analyze the graded structures in terms of suitable kinematic model
that incorporates the realistic structural response. Hence in the present work prominence
has been given to study the structural response of FGM skew plates/shells and sandwich
plates/shells under linear and non-linear analyses.
The following interpretations have been drawn from the literature study.
 FGM structures are generally characterized by their spatial variation of material
properties in the predetermined fashion to achieve the desirable material properties in
the chosen direction. The primary intention behind the development of these types of
materials is to sustain large temperatures and high temperature gradients. Due to their
anti-symmetric nature of properties with respect to middle plane, bending-stretching
coupling is anticipated. In order to capture the realistic variation of shear deformation,
a suitable higher order theory should be employed in the analysis.
 In heterogeneous materials like FGM, selection of appropriate homogenization
scheme plays important role in defining the material properties in proportion to their
volume fraction values. The choice of proper homogenization approach should be
based on the gradation relative to the extent of a typical representative volume element
(RVE). If the graded material properties are relatively slow-changing functions of
spatial coordinates, standard homogenization methods that disregard the
heterogeneous nature of material property at global level can be accomplished. In this
regard, the averaging techniques like Mori-Tanaka and self consistent schemes which
77
include the interaction among the neighboring inclusions may be the better options.
Most of the literatures incorporate the variation of material properties by means of
simple rule of mixture; while very few literatures are concerned with the averaging
methods.
 FGM materials are mainly developed to serve the purpose in high temperature
environments, thus necessitate the accurate prediction of thermal distribution in the
structures. Hence most of the studies performed in the literature consider the thermal
profile in linear and non-linear form with temperature dependent material properties.
But the studies on FGM skew plates/shells under thermal environment based on higher
order model seem to be missing in the body of literature.
 General FGM structures undergo failure due to buckling, large amplitude deflections
and excessive stresses caused by the thermal and combined thermo-mechanical
loading. In such case, the strain-displacement equations should be incorporated in the
non-linear sense. To analyze the FGM structures under large deflection with small
strains and moderate rotations, the non-linear strain model was considered in Green-
Lagrange sense by many researchers. All the studies presented so far employs the
assumption of von Kármán to obtain the final form of strain equations. Further, when
the non-linearity is severe, it becomes mandatory to consider all the non-linear terms
appear in the Green-Lagrange equations. Only few studies are reported on the
geometric non-linear analysis of FGM plates and shells and no literature has been
located regarding geometric non-linear analysis of FGM sandwich plates/shells.
 FGM plate/shell structure show reserve strength after the critical load is reached which
was described as post buckling strength. Further, the sudden change of equilibrium
from one state to another involving large deformation should be investigated properly
from design point of view. In addition, the variation of material properties makes the
analysis more complicated. Keeping these aspects in view, few studies are performed
to address the post buckling behavior of FGM plates and shells. Again the availability
of literature document to address the post buckling response of FGM sandwich plates
through proper higher order kinematic model was missing in the literature.
In order the overcome the drawbacks that are elaborated in Section 2.7; a brief
framework of the present investigation was discussed here. From the short discussion made
on published literatures, it was manifested to propose an appropriate model to analyze the
FGM skew single layer/sandwich plate/shell under thermo-mechanical loading. To
incorporate the non-linear nature of geometric stiffness matrix associated with buckling
78
phenomenon, it is significant to consider all the non-linear terms in the strain model for
post-buckling analysis. Finally, a non-linear model that accurately predicts the transverse
displacement variation by means of quadratic thickness term was accomplished in the
present analysis.
Various features that are accommodated in the present research are presented
below.
 A higher order displacement based model that accommodate the cubic and quadratic
variation of thickness term, respectively, in the in-plane and transverse displacement
field has been employed to accurately predict the bending stretching coupling exists in
FGM structures. Such a model includes the normal strain and its derivative in the
kinematics thus overall response of the plate can be efficiently accomplished.
 A higher order non-linear model was assumed for geometric non-linear and post
buckling analyses of sandwich plates/shells by employing C0 formulation.
 Two kinds of homogenization approaches viz., Mori-Tanaka and rule of mixture are
employed and comparison statement has been drawn for FGM skew plates based on
results obtained from these methods.
 To accurately predict the non-linear response of FGM structures, all the non-linear
terms present in the Green-Lagrange equations are incorporated in the non-linear finite
element formulation.
 A suitable finite element code has been developed in FORTRAN 90 environment for
static, dynamic and stability analyses of FGM skew plates. In addition, a more versatile
numerical code in MATLAB (R2013b) platform was also developed to perform
linear/non-linear analysis of sandwich plates/shells.
 Finally, a wide range of numerical problems are solved in the framework of developed
codes that dictate the response of FGM plate/shell structures considering linear and non-
linear strain-displacement relations.
2.9 SUMMARY
In this chapter, an extensive research works performed on FGM plates/shells by
considering linear/non-linear strain-displacement relations are studied in detail. In
particular, works related to static, free/forced vibration, and buckling responses of FGM
plates/shells are discussed while giving a brief explanation about the methodology and
displacement model incorporated for the analysis. Various important conclusions that have
been arrived from different studies are also discussed. In addition, literatures available on
79
static, free vibration and buckling analyses of FGM sandwich plates are studied. Finally,
literature studies on geometric nonlinearity and post buckling analyses of FGM
plates/shells have been discussed in detail. From the brief literature survey carried out in
Chapter 2, research gaps are identified which formed the basis for the present research
work. In the end, a summary of the important research works that serves as background for
the present research work is provided in Table 2.1.
80
Table 2.1: Summary of the important research works
A summary of the important literature papers discussed in Chapter 2 is provided in the table below.
S. No Author Model implemented Major Contributions

Shear Deformation Theories
The proposed model has cubic variation of thickness
Higher order displacement model is proposed
1 Reddy (1984) term in in-plane field and constant variation of
for laminated plates.
transverse displacement component.
C0 finite elements are formulated based on Higher order models having 7, 9, 11 and 12 nodal
Manjunatha and Kant
2 different higher order shear deformation unknowns are assumed for sixteen and nine noded
(1992)
theories. Lagrangian elements.
Higher order model proposed by Manjunatha Static, free vibration of FGM plates is performed by
Talha and Singh and Kant (1992) is implemented for the analysis considering the cubic and quadratic variation of in-
3
(2010) of functionally graded plates. plane and transverse displacement components,
respectively.
Displacement models are based on CPT, FSDT, Static analysis of FGM sandwich plates is performed
sinusoidal, and TSDT by appropriate by considering various types of sandwich plates.
4 Zenkour (2005)
substitution of higher order terms in the
displacement field.
Hyperbolic sine terms are incorporated in the in- Displacement model is implemented to perform static
plane fields in addition to Murakami’s zig-zag analysis of functionally graded sandwich plates.
5 Neves et al. (2012)
function. Quadratic variation of transverse
displacement component is assumed.
81
Analysis of Skew plates/shells
1 Butalia et al. (1990) Nine-node quadrilateral Heterosis element based Serendipity and Lagrangian elements are used in the
on Mindlin formulation has been employed. analysis.
2 Reddy and High precision traiangular element has been Simply supported and clamped antisymmetric angle
Palaninathan (1999) employed for free vibration of skew laminates. ply laminates are analyzed under different skew
angles.
3 Babu and Kant (1999) The FSDT and HSDT based finite element Laminated composite skew plates and sandwich
models are proposed for stability analysis of panels are treated by considering different lamination
skew laminates. schemes.
4 Ganapathi et al. FSDT based model is assumed in conjunction Buckling analysis of functionally graded skew plates
(2006) with finite element method. is performed.
5 Ganapathi and Finite element method based on FSDT based Thermal buckling of FGM skew plates is performed
Prakash (2006) model is employed. by considering various skew angles.
Geometric nonlinear analysis of FGM plates/shells
Displacement field and shallow shell element Load-deflection curves, principal bending and
Srinivasan and Bobby based on Cowper et al. (1970) has been membrane stresses are plotted for different support
1
(1976) extended to study the non-linear response of conditions by incorporating various skew angle
skew plates. values.
Linear, Serendipity, Lagrangian and Heterosis
Mindlin plate theory in conjunction with finite elements are used to obtain the load-displacement
2 Pica et al. (1980)
element method is employed in the analysis. response of square, skew, circular and elliptical
plates.
Praveen and Reddy Higher order theory of Reddy (1984a) is Functionally graded plates are analyzed under
3
(1998) implemented by incorporating von Kármán mechanical loading and various thermal conditions
82
assumptions. prescribed on the top and bottom surfaces.
Arciniega and Reddy Improved version of FSDT is combined with Isotropic, laminated composites and FGM shells are
4
(2007a) finite element method. analyzed under non-linear response.
Functionally graded plates are analyzed under thermal
Zhao and Liew Displacement model is based on FSDT in
5 and mechanical loading for different combinations of
(2009b) conjunction with Petrov-Galerkin approach.
material and geometric parameters.
Zhao and Liew (2009 FSDT based displacement model is assumed in Functionally graded shells are treated under
6
c) the frame work of Petrov-Galerkin method. mechanical and thermal loading conditions.
Post buckling analysis of plates/shells
Post buckling of cylindrical, spherical, circular and
Mindlin shallow shell formulation is used in the
1 Pica and Wood (1980) square plates is performed under in-plane shear and
framework of finite element method.
axial loading conditions.
Temperature dependent properties and plate
Semi analytical approach based on CPT is
2 Yang and Shen (2003) foundation interaction effects are incorporated in the
performed.
analysis.
Mathematical model is based on FSDT and von Analytical solution is obtained for functionally graded
3 Wu et al. (2007)
Kármán strains. plates subjected to mechanical and thermal loading.
Post buckling analysis of functionally graded
FSDT model combined with element free kp-
4 Liew et al. (2012) cylindrical shells is carried out under thermal load and
Ritz method has been employed.
axial compression.
Functionally graded plates are analyzed under thermal
C0 based HSDT model is used in conjunction
5 Lal et al. (2013) and mechanical loading by considering randomness of
with finite element method.
material properties
83
CHAPTER 3
IDEALIZATION, MATHEMATICAL MODELLING AND
FE FORMULATION OF FGM PLATES AND SHELLS
3.1 INTRODUCTION
Although the concept of functional grading of mechanical/thermal properties in the chosen
direction appears to be an engineering invention, the concept is not novel. This graded
concept has been occurring in nature in different forms. For example, according to the
depth and location in the body, certain toughness and elastic qualities are graded in the
human skin. An engineering definition generally implies the grouping of two distinct
isotropic material phases which offer various superior properties that serve the functional
requirement of the structure in efficient way. Although, several number of combinations
can be accommodated in the concept, advanced ceramic materials such as zirconia,
alumina, silicon carbide and tungsten carbide; and engineering alloy of magnesium,
aluminium, copper, steel, tungsten are given the preference in the list. To describe in
simple definition, in the concept of FGM, ceramic material and engineering alloy are
combined together to achieve an optimum graded profile configuration in the desired
direction. The output will be a smooth and continuous variation of microstructure with the
properties of high toughness and thermal resistance. Such graded concept has
demonstrated that compositional micro/macrostructure can not only eliminate undesirable
effects such as stress concentration, but can also generate unique positive function (Gasik
2010). A general configuration of FGM comprise of large concentration of ceramic
material at the high temperature exposed side, while large concentrations of metal
component are placed at regions where mechanical strength is the vital parameter.
Primarily, the concept of FGM has emerged to serve in high thermal related structures.
Since then, FGMs have made their foot print in heaps of applications such as nuclear
reactors (Igari et al. 1990), dental and medical implants (Oonishi et al. 1994), piezoelectric
and thermo electric devices (Tani and Liu 1993; Osaka 1990; Watanabe 1993), fire
retardant doors (Getto and Ishihara 1996), energy sources, optical, civil, automotive and
ship building industries. With the wide range of application areas, FGMs becomes
promising candidates for future intelligent composites in fast computers and aerospace
84
industry, environmental sensors and also to form macro and nano devices (Moya 1995).
The most employed fabrication technique to combine the distinct materials with varying
volume fraction is powder metallurgy (Ahu et al. 1996). Apart from this technique several
other techniques are also located in the literature (Yamoka et al. 1993; Fukui 1991).
Having possessing all these features, the accurate modeling and optimum design of FGM
structures is the challenging task for designers to meet the various practical needs in
effective manner.
The present chapter is organized in the following fashion. It starts with the brief
explanation about the various homogenization schemes and micromechanical models
required for proper idealization of FGMs. Subsequently, the geometrical configuration of
single layer and sandwich FGM plate/shell is presented along with the expressions for
different power law describing functions. Followed by this topic, material modeling, skew
boundary transformation and solution for heat conduction equation are presented. Having
defined the configuration and material modeling, the FE formulation that includes
kinematics field by considering constant/quadratic variation of transverse displacement has
been presented with the element description. Next, the governing equations obtained for
static, dynamic and buckling analyses are discussed. The procedure for geometric non-
linear and post buckling problem is discussed under succeeding subsections. In the end,
various support conditions and computer implementation are included followed by the
summary report of this chapter.
3.2 HOMOGENIZATION OF FGMS

A successful implementation of any engineering concept in to practical application relies
on the assumption of proper micromechanics model. Since FGMs are highly
heterogeneous in nature, having smooth and graded variation of material properties in the
required direction, suitable choice of homogenization approach should be accomplished for
realistic structural modeling. Various types of approaches available in the literature for
homogenization of FGMs are discussed below.
3.2.1 Self consistent scheme
This approach was found more suitable for ellipsoidal inclusion embedded in a matrix
possessing the effective mechanical properties of the composites (Hill 1965; Hashin 1968;
Bhaskar and Vardhan 2001). The model accurately defines the each reinforcement
inclusion embedded in a continuum material whose properties are similar to those of the
composites. The same modulus was predicted in the composite material where the role of
85
the two phases gets interchanged and it does not distinguish between the reinforcement and
matrix phase. Hence, this approach will be more suitable for the composite having
interconnected skeletal microstructure.
3.2.2 Mori-Tanaka approach
If the composite has regions of the graded microstructure with a well defined continuous
matrix and a discontinuous particle phase, this method yield better results (Mori and
Tanaka 1973; Benvensite 1987). The assumption of spherical particles embedded in a
matrix was considered. The primary matrix phase is assumed to be reinforced by spherical
particles of secondary phase. Suppose Vc and Vm represent the volume fraction of ceramic
and metal, respectively, then the volumetric relation Vc+Vm=1.0 should be satisfied.
According to the Mori-Tanaka scheme (Mori and Tanaka 1973) the bulk modulus
(B), shear modulus (G), modulus of Elasticity (E), Poisson’s ratio (γ), heat conductivity
coefficient (k), coefficient of thermal expansion (α) of the composite represented by
corresponding ceramic and metal properties with subscripts ‘c’ and ‘m’ may be expressed
as,
B  Bm Vc

Bc  Bm 3( Bc  Bm )
1  (1  Vc )
3Bm  4Gm
G  Gm Vc G (9 Bm  8Gm )
 , where f1  m
Gc  Gm 3(Gc  Gm ) 6( Bm  2Gm )
1  (1  Vc )
3Gm  f1
9GB 3B  2G
E , 
3B  G 6 B  2G (3.2.1)
k  km Vc

kc  k m (k  k )
1  (1  Vc ) c m
3km
c  m  1
1  Vc Vc 
   m (1  Vc )   cVc     
1 1 B Bm Bc 

Bc Bm
3.2.3 Composite sphere assemblage model

In this method, the effective mechanical properties of the composites are estimated using
analytical method. It was presumed that the composite material is filled with a fractal
assemblage of spheres that embedded in a concentric spherical matrix having different
diameters (Hashin 1962; Hashin 1964). In this sense, the spheres will completely fill the
volume of the composites.
86
3.2.4 Composite cylindrical assemblage model
This method will be more appropriate if both the matrix and reinforcing phase are isotropic
in nature and can be applied to orthotropic composites (Hashin and Rosen 1964; Hashin
1979). In the model, the microstructure defined by representative volume element (RVE) is
transversely isotropic in planes perpendicular to the orientation of fibers.
3.2.5 The simplified strength of materials method
This method offer easy implementation of the concept and increased computational
efficiency. In this approach, the matrix phase was reinforced with a periodic array of
square fibers (Chamis and Sendeckyj 1968; Gibson 1994). Hence, this method has proven
to be more benefit to calculate the orthotropic strength of fiber reinforced composite
laminate from the strength properties of matrix and fiber and corresponding volume
fractions.
3.2.6 The method of cells
This method is identical to that of Chami’s method (Chamis and Sendeckyj 1968) of
strength of materials, but it was computationally rigorous since it assumes a RVE that
involves a larger volume proportion of matrix phase (Aboudi 1991).
3.2.7 Micromechanical models
These models are more applicable for both isotropic and orthotropic composite materials
and RVE of this model can be constructed via FE simulations. By incorporating the
different material combinations, curve fitting data may be formed by FE simulation (Reiter
et al. 1997; Caruso and Charnis 1986)). Since this method involves the modeling of three-
dimensional finite elements, it is considered to be more accurate compared to other
homogenization methods. The major drawback is the multiple modeling of RVEs in order
to determine the effective properties corresponding to the different volume fraction
constituents. But such an issue can be alleviated by making use of proper computer
software that can accommodate the multiple modeling concepts.
3.3 IDEALIZATION OF FGMS
A general FGM structure is characterized by its highly heterogeneous nature of material
properties with a prescribed distribution of volume fractions of the constituent phases. For
particular analysis, it seems to be functional to idealize them as continua with smooth
gradual variation of material properties in the spatial coordinates. Hence, it becomes
essential to incorporate the proper micromechanical model that can characterize the
distribution of material properties of system in accurate sense. Two types of gradation
87
methods are popular in the previous literatures and therefore widely applied to solve many
problems.
3.3.1 The Exponential law
Idealizing the FGM structure by means of exponential law variation is very common in
fracture mechanics studies (Suresh and Mortensen 1998). If Pc and Pm represents the
properties of top and bottom surfaces of the plate/shell, respectively, then the resultant
effective mechanical property of a composite at a distance ‘z’ from the reference surface is
given by
  2z   1 P 
P  z   Pc exp   1    , where   ln  c  (3.3.1)
  h  2  Pm 
3.3.2 The Power law

This form of representing the material variation was generally applied in stress analysis of
FGM structures (Suresh and Mortensen 1998). Considering the identical notation for
material properties as in the case of exponential law, the property of composite is given by
n
 z 1
P  z    Pc  Pm      Pm (3.3.2)
h 2
In Eq. (3.3.1) and (3.3.2), the parameters  and n appears in the expression defines the
grading of the material dispersed in the structural geometry.
3.4 GEOMETRICAL CONFIGURATION

3.4.1 Single layer FGM plate/shell
A single layered FGM plate of length a, width b and thickness h is depicted in Figure 3.1.
The upper layer of the plate/shell surface is ceramic rich; while the bottom portion of the
plate/shell surface is metal rich with a gradation zone having smooth variation of material
properties (either thermal/mechanical) in between the two surfaces. Even though, the
gradation concept can be attained in all the three directions (x, y and z), the present analysis
focused on 1D FGM problems i.e., graded properties are incorporated in the thickness
direction only.
In the present research work cylindrical, spherical, hypar and hyperboloid shell
forms are analyzed under bending, vibration and buckling analyses. The formulation
incorporates all the three radii of curvature in the mathematical part thus leading to a more
generalized formulation for FGM shell problems. Shallow shell assumptions are
implemented in the program.
88
z
Ceramic rich
rich
Transition x h
zone
Metal rich
a
Figure 3.1 A Schematic representing single layer FGM plate
Reference plane
Ceramic rich
b
R= ∞
h Metal rich
x R= Rx
a
Figure 3.2 A Schematic representing single layer FGM cylindrical shell
A typical sketch representing the FGM cylindrical shell panel having dimensions a x b x h
and radius of curvature Rx in x direction is presented in Figure 3.2. Since material profile
in FGMs is represented by their volume fraction distribution of constituents in the
89
thickness direction, the following relationship is incorporated to estimate the volume
fraction of ceramic (Reddy 2000; Praveen and Reddy 1998; Reddy and Chin 1998).
n
 z
Vc   0.5   (3.4.1)
 h
where Vc represents the volume fraction of ceramic and the exponent term n appears in the
above expression is generally termed as volume fraction index, gradient index, power law
exponent and material profile parameter. To avoid confusion, the term volume fraction
index is used throughout the thesis report. The parameter volume fraction index defines the
composition gradation of the material in the thickness direction. The value of n=0,
represent the pure ceramic plate; while n=∞, represent the pure metal plate and the value in
between represent the gradated plate representing the spatial variation of properties
according to Eq. (3.3.2). However, for all cases, the volumetric relationship Vc+Vm=1.0
should be satisfied. To visualize the variation of volume fraction of ceramic through-the-
thickness for single layer FGM plate/shell, the plot is presented in Figure 3.3.
0.6 n=0.2
n=0.4
0.4
n=0.5
Non-dimensional depth (z/h)
n=0.8
n=1
0.2 n=1.5
n=2
n=2.5
0.0 n=2.8
n=3
n=4
-0.2
n=5
n=10
-0.4 n=20
n=30
n=50
-0.6
0.0 0.2 0.4 0.6 0.8 1.0
Volume fraction of ceramic (Vc)
Figure 3.3 Through-the-thickness variation of volume fraction of ceramic (Vc) for

different n values
From Figure 3.3, one can visualize the variation of Vc across the thickness of the
plate/shell geometry for different values of volume fraction index. At the top surface of the
plate/shell, Vc=1.0 represents the pure ceramic and at the bottom surface of the plate/shell
Vc=0 represents the pure metal, which is the observation similar to the one explained with
the help of Eq. (3.3.2). Among the different cases of volume fraction index considered, the
linear variation of Vc is observed for the case of n=1.0. As the value of volume fraction
90
index becomes higher, the concentration of ceramic decreases while concentration of metal
increases. Further, if the volume fraction index attains the value greater than 10, significant
reduction in the ceramic content is observed towards the bottom of the plate.
Knowing the distribution of material constituents in terms of volume fraction
parameter, the effective properties of the composite at any height ‘z’ can be estimated by
incorporating the following rule of mixture (Suresh and Mortensen 1998).
P( z)  (Pc  Pm ) Vc  Pm (3.4.2)
The effective property P(z) appear in Eq. (3.4.2) may represent the mechanical and
thermal properties of the plate/shell under consideration. The mechanical properties
include Young’s modulus, Poisson’s ratio and density of the ceramic and metal
components; while thermal properties include co-efficient of thermal expansion and
thermal conductivity for ceramic and metal components. In most of the cases, the variation
of Poisson’s ratio is neglected, since the mechanical entities (deflection and stresses) are
least affected (Delale and Erdogan 1983; Chi and Chung 2006) by its value. The variation
of Young’s modulus (E) and thermal expansion (𝛼) through the thickness for single layer
stainless steel (SUS3O4)/silicon nitride (Si3N4) constituted FGM is demonstrated in Figures
3.4 and 3.5, respectively. As the n value approaches the bottom portion of the plate/shell,
the stiffness of the plate reduces considerably, and linear variation of Young’s modulus
and thermal expansion co-efficient is discerned for n=1.0.
0.6 n=0.2
n=0.4
n=0.5
0.4
n=0.8
n=1
0.2
n=1.5
n=2
n=2.5
0.0 n=2.8
n=3
n=4
-0.2 n=5
n=10
n=20
-0.4
n=30
n=50
-0.6
2.00E+011 2.40E+011 2.80E+011 3.20E+011
Young's modulus (E)
Figure 3.4 Through-the-thickness variation of Young’s modulus (E) for different n

values
91
0.6
n=0.2
n=0.4

0.4 n=0.5
n=0.8
n=1
0.2 n=1.5
n=2
n=2.5
0.0
n=2.8
n=3
-0.2 n=4
n=5
n=10
-0.4 n=20
n=30
n=50
-0.6
0.000008 0.000010 0.000012 0.000014 0.000016
Co-efficient of thermal expansion ()
Figure 3.5 Through-the-thickness variation of thermal expansion (𝛂) for different n

values
3.4.2 Sandwich FGM plate/shell
In case of sandwich construction offered by laminated composites, various problems such
as delamination and high stress concentration may occur, owing to large jump of material
properties at the layer interfaces (Drake et al. 1993; Lee and Erdogan 1995; Praveen and
Reddy 1998; Chung and Chi 2001). To address these issues, over the course of time, the
concept of functional grading has been employed in sandwich structures. The smooth and
gradual variation of material properties from point to point ensures no interfaces in FGM,
so that any large variation of material properties could be avoided in effective way. In
sandwich FGMs, modeling of plate/shell geometry could be exploited in two alternative
ways. In the former case, the top and bottom layers are occupied by the ceramic and metal,
respectively, thereby the effective properties in the core portion obey the rule of graded
distribution of the constituent materials. In the later case, the homogenous ceramic core is
embedded between the two layers of metal constituent; thereby the bottom and top layer
utilizes the advantage of FGM. The rule of mixture to be followed by the material
constituents in both the cases is governed by the volume fraction of the constituent
materials (Zenkour 2006). Such type of material modeling could also be exploited in case
of smart structures where the actuators and sensors are placed in the face layers by
appropriately selecting the core part ((Pan and Han 2005; Xiang et al 2010; Alibeigloo
2010; Loja et al. 2013).
92
It is worth to mention that, there exist no interface between the core and top/bottom
layers as the mechanical/thermal properties have smooth and continuous variation, and
hence the possibility of delamination and other failure modes encountered in conventional
sandwich construction could be avoided. In the present research work, modeling of
functionally graded sandwich plate (FGSP) is carried out in two alternative ways with
respect to the accomplishment of isotropic/FGM material in the core/face layers. In both
the cases, the FGSP is separated into three microscopically heterogeneous layers, where
the thickness of each layer is defined with respect to the mid-plane of the plate (i.e, z = 0).
For the sake of numerical analysis, thickness of all the layers are represented from the
bottom of the plate (z = -h/2) in both the types of modeling. The term Vc may assume
appropriate expression depending upon the type of material distribution exists at every
layer of the plate/shell configuration.
3.4.2.1 Sandwich FGM plate/shell with FGM skin

In this type, the plate/shell geometry is modeled such that the core portion is
defined by the ceramic material, wherein the top and bottom portions are defined by metal
constituent (Figure 3.6). By such an accommodation, the top and bottom layers of the
sandwich plate exhibits smooth graded distribution of material properties and the core
layer represent the isotropic property.
h3 METAL
FGM SKIN
h2 ISOTROPIC CORE
CERAMIC
h X
h1 (2) CERAMIC
FGM SKIN
h0 METAL
Figure 3.6 Geometry of FGM sandwich having graded skin layers and isotropic core
Since, the thicknesses of three layers may vary with respect to the neutral or
reference plane; the plate may display the symmetric or anti-symmetric nature of material
properties. For the sake of ease understanding, a simple notation is used to refer the kind of
sandwich plate to be considered for the analysis. For instance, a notation of 1-2-1 (i.e.,
bottom layer-core-top layer thickness) is used to symbolize that the top and bottom
93
segments have identical thickness values, whereas the core thickness equals twice the
bottom/top thickness. Since, the variation of material properties in each layer is not
analogous; the expression for Vc at each layer is approximated by the suitable exponential
function (Jin and Batra 1996). The expression for Vc related to each layer is defined by
means of following expression.
n n
 z  h0   z  h3 
Vc (1)
  ;Vc  1.0;Vc  
(2) (3)
 (3.4.3)
 h1  h0   h2  h3 
0.6 0.6
0.5 0.5
0.4 0.4

0.3 0.3
0.2 0.2
0.1 0.1
n=0.2,, n=0.5, n=1 n=0.2,, n=0.5, n=1
0.0 n=2, n=5, n=10 0.0 n=2, n=5, n=10
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
Volume fraction of ceramic (Vc) Volume fraction of ceramic (Vc)
(a) 1-1-1 thickness scheme (b) 2-1-2 thickness scheme

0.6
0.5
0.4
0.3
0.2 n=0.2
0.1
n=0.5
n=1
0.0 n=2
-0.1
n=5
n=10
-0.2
-0.3
-0.4
-0.5
-0.6
0.0 0.2 0.4 0.6 0.8 1.0
(c) 1-2-1 thickness scheme
n=0.2
n=0.5
n=1.0
Figure 3.7 Through-the-thickness
n=2.0
variation of volume fraction of ceramic (Vc) for
n=5.0
different types of sandwich plates having FGM skin
n=10.0
2-1-2 type plate
94
Depending on the layer identity, the term Vc appear in Eq. (3.4.2) to calculate the
effective material property is substituted by the appropriate expression from Eq. (3.4.3).
The diagram representing the variation of Vc through the thickness for three types of
sandwich schemes is depicted in Figure 3.7. In all cases, the linear variation of Vc is
ensured at core due to the concentration of isotropic ceramic material and further the depth
of concentration depends upon the thickness of core chosen with respect to the total
thickness of the plate. As assumed the volume fraction of ceramic at top and bottom
surfaces becomes zero, thus representing the higher concentration isotropic metal. Also,
when n=1.0, the linear variation of graded properties at the bottom and top face layers is
ensured.
3.4.2.2 Sandwich FGM plate/shell with FGM core
In this approach, the bottom and top layers of the plate/shell panel are characterized by
metal and ceramic part, respectively. Hence, a sandwich plate/shell having isotropic face
sheets and core having smooth and graded variation of material properties could be
modeled. By such option, minimum jump in the material properties and stresses at the
layer interfaces can be ensured.
Z
h3
CERAMIC
ISOTROPIC SKIN
h2 CERAMIC
FGM CORE X
h
h1 METAL
ISOTROPIC SKIN
h0 METAL
Figure 3.8 Geometry of FGM sandwich having graded core and isotropic face sheets
The volume fraction Vc at each layer of sandwich FGM is described by the
following expression.
n
1 z 
Vc
(1)
 0;V
c
(2)
   c  ;Vc(3)  1.0 (3.4.4)
 2 hc 
Since the bottom segment is represented by pure isotropic material (metal), the
value of Vc for the bottom layer is zero. For core layer, the material profile distribution is
defined by FGM concept and hence the volume concentration similar to the one described
in Eq. (3.4.1) is used to evaluate the effective material properties. Further, the top segment
of the sandwich plate is represented by another pair of isotropic material (ceramic) and
95
hence conquers the value equal to unity. In Figure 3.9, the variation profile of Vc for
different sandwich plates having FGM skin is presented. The top and bottom layers of
plate/shell ensures linear variation of volume fraction and its value become unity for upper
layer and zero for lower layer depending on the type of material distribution (either
ceramic or metal). Among the different types of plates considered for the illustration, the
plate with 1-2-1 thickness scheme exhibit parabolic variation of Vc in core due to the
higher thickness of core compared to other layers. A contradictory observation can be
discerned for core having thickness about 1/5 times the total plate thickness (2-1-2 plate).
Hence, a graphical visualization of Vc through-the-thickness may allow the designer to
arrive for optimum configuration in terms of design requirements.
0.6 n=0.2
0.6
n=0.5
0.5 n=1 0.5 n=0.2
n=2 n=0.5
0.4 0.4 n=1

n=5 Non-dimensional depth (z/h)
0.3 n=10 0.3 n=2
n=5
0.2 0.2 n=10
0.1 0.1
0.0 0.0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
0.6
0.5
0.4
0.3
0.2
n=0.2
0.1 n=0.5
0.0 n=1
n=2
-0.1
n=5
-0.2 n=10
-0.3
-0.4
-0.5
-0.6
0.0 0.2 0.4 0.6 0.8 1.0
different types of sandwich plates having FGM core
96
3.5 FOUR-PARAMETER POWER LAW DISTRIBUTION
0.6 0.6
0.5 0.5
0.4 0.4

0.3 0.3
0.2 0.2
0.1 0.1
0.0 0.0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
(a) a=1,b=c=0 (b) a=1, b=1, c=2

0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0.0 0.0
z/h
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0
(c) a=1, b=1, c=4 (d) a=1, b=0.5, c=2
0.6
0.5
0.4
0.3
0.2
0.1
0.0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
0.0 0.2 0.4 0.6 0.8 1.0
(e) a=0.8, b=0.2, c=3

single layer FGM plate/shell according to four-parameter power law distribution
97
The term Vc defined in Eq. (3.4.1) that dictates the material profile variation of FGM in the
thickness direction leads to conventional form of FGM i.e., the top portion will have high
ceramic content and bottom portion will have high metal content. But in real time
applications the need arises where the designer has to be opt for the FGM profile other
than the conventional one. For example, either the top/bottom segment may have
ceramic/metal rich layer, or the top portion may have 50% ceramic and 50% metal or both
top and bottom may have ceramic/metal constituent.
To achieve such profile, a FGM distribution that dictated by four parameters in the
expression has been proposed in the following manner.
n
  z  z1
c
FGM I / FGM II ( a1 / b1 / c1 / n ) : Vc  1  a1  0.5    b1  0.5   

  h  h  
(3.5.1)

where the different parameters a1, b1 and c1 dictate the material profile variation through
the thickness and exponent n is assumed to vary at every point to represent the different
types of plate configurations. As per Eq. (3.5.1), each FGM profile leads to distinct plate
configuration, but the deviation among the results (either deflection or frequency) based on
different configuration was observed after the fourth decimal place only (Tornabene 2009).
With the intention to present significant results, the author opted first configuration
(positive sign for inner bracket terms) to perform the different analyses. The user can
assign different values to the parameters (a1, b1, c1 and n) to obtain the material
configuration having symmetric and asymmetric profiles. However, the values chosen for
the four parameter parameters should satisfy the volumetric relationship among the two
constituents i.e., Vc+Vm=1.0.
3.6 TRANSFORMATION OF SKEW BOUNDARY

A FGM skew plate having the overall dimensions a x b x h m and skew angle  is
depicted in Figure 3.11. In case of skew plate/shell, one of the skew boundary of the plate
is not parallel to the global axis system (y) , and thus it is necessary to transform the nodal
unknowns of the skew edges from global axes (x and y ) to local axes (x′ and y′) at the
element level by means of suitable transformation rule. Then the transformed matrices
formed at the element level are assembled to form the global matrices using the standard
assembly procedure of finite element analysis. For the elements which are not lying on the
skew edges, the transformation procedure is not mandatory. The transformation matrix [T]
98
utilized for the skew boundary for the case of constant and quadratic variation of
transverse displacement component is given in Appendix A.
y y’

b
h FGM x, x’
Figure 3.11 Geometry of single layer FGM skew plate/shell
3.7 THERMAL ANALYSIS

In the present investigation, the temperature field is assumed to be constant in the x-y plane
and has non-linear variation in the thickness direction (z). The temperature-dependent
material properties of constituents (metal and ceramic) are considered for different
analyses. If the thermal dependent properties of two constituents forming FGM geometry
are known, then the temperature dependent property of composite can be represented as
(Touloukian 1967)
P1
P  P0 (  1  PT
1  P2T  PT
2
3
3
) (3.7.1)
T
where P0, P-1, P1, P2, P3 are the temperature coefficients T (K) and assume the
value of constants in the cubic fit of Eq. (3.7.1). The temperature through the thickness is
represented by the steady state one-dimensional Fourier equation of heat conduction as
expressed below.
99
d  dT 
 k ( z) 0
dz 
(3.7.2)
dz 
In order to solve Eq. (3.7.2), the following thermal boundary conditions are
imposed on the top and bottom surfaces of the plate/shell.
T  Tc at z  h / 2and T  Tm at z   h / 2 (3.7.3)
where Tc and Tm denote the temperature of ceramic and metal, respectively.
Further, the solution for Eq. (3.7.2) may be obtained by means of polynomial series as
T ( z )  Tm  (Tc  Tm ) ( z, h) (3.7.4)
where
 2 z  h  kcm  2 z  h 
k 1
k 2cm  2z  h 
2 k 1
k 3cm  2z  h  
3 k 1
           
1  2h  (k  1)km  2h  (2k  1)k 2 m  2h  (3k  1)k 3m  2h  
 ( z , h)   4 k 1 5 k 1 
C
 k 4 cm  2z  h  k 5cm  2z  h  

 (4k  1)k 4 m  2h  
(5k  1)k 5 m  2h 
 
where
kcm k 2cm k 3cm k 4 cm k 5cm
C 1      and kcm  kc  km
(k  1)km (2k  1)k 2 m (3k  1)k 3m (4k  1)k 4 m (5k  1) k 5 m
3.8 ASSUMPTIONS
Different mathematical formulations presented in this chapter are based on the following
assumptions.
 The mid plane of the plate/shell is taken as reference plane.
 The thermo-mechanical properties of the plate/shell vary according to rule of
mixture in thickness direction only.
 The temperature field is assumed to be applied in the thickness direction and
constant in the in-plane direction.
 The thickness of the plate/shell is constant.
 The shell is assumed to be shallow.
 The kinematics filed may have cubic variation of thickness term in the in-plane part
and constant/quadratic variation of thickness in the transverse displacement.
3.9 MATERIAL MODELLING
3.9.1 Constitutive relationship
The elastic stress-strain relationship is based on the assumption that whether the problem is
plane stress (ϵzz=0) or not (ϵzz≠0). If the constant variation of transverse displacement is
100
incorporated in the assumed kinematic expression, the general constitutive relationship
including the effect of thermal field may be expressed as
 xx   Q11 Q12 0 0 0    xx  1  
      
 yy  Q21 Q22 0 0 0    yy  1  
     
 yz    0 0 Q33 0 0    yz   0    z, T  T 
 (3.9.1)
   0 
0    xz  0  
 xz  
0 0 Q44
     
 xy   0 0 0 0 Q55    xy  0 
    
E( z)  E( z) E( z)
where Q11  Q22  , Q12  Q21  , Q33  Q44  Q55  are the stiffness
(1   )
2
(1   )
2
2(1   )
coefficients depends on the Young’s modulus (E) and Poisson’s ratio (γ) of the material. In
addition, thermal expansion co-efficient (α) is a function of depth (z) and temperature (T)
of the material.
Alternatively, if the effect of normal strain is incorporated in the analysis, the general
constitutive relationship including the thermal field may be represented as
 xx   Q11 Q12 Q13 0 0 0    xx  1  
        
 yy  Q21 Q22 Q23 0 0 0    yy  1
  
 zz   Q Q Q33 0 0 0    zz  0  
(3.9.2)
       
    

13 23
z , T T
  yz   0 0 0 Q44 0 0    yz  0 
   0  
0 0 0 Q55 0    xz  0 
 xz       
  xy   0 0 0 0 0 Q66    xy  0 
  
where the stiffness coefficient Qij contains the terms elastic modulus (E) and Poisson’s
ratio (), in which only the term E vary as a function of depth (z). Hence, homogenization
scheme can be applied only to Young’s modulus (E) and Poisson’s ratio () is assumed as
constant.
E ( z )(1   2 )  E ( z )(1   ) E( z)
Q11  Q22  Q33  , Q12  Q13  Q23  , Q44  Q55  Q66 
(1  3  2 )
2 3
(1  3  2 )
2 3
2(1   )
By computing the values of Young’s modulus at any height (z) of the sandwich panel from
Eq. (3.4.2), it becomes simple procedure to calculate the stiffness co-efficient Qij at x-y-z
axes system.
3.9.2 Element description
In order to model the geometry of FGM plate/shell, a nine node C0 isoparametric
Lagrangian element is utilized in the present investigation (Figure 3.12).
101

7 6 5
8 9 4 ξ
1 2 3
Figure 3.12 Isoparametric nine node Lagrangian element in natural co-

ordinate system
For the assumed nine node element the expressions for interpolation/ shape functions Ni
which are functions of natural co-ordinates ( and ) are described below.

For corner nodes:
1 1 1 1
N1    1  1  , N3    1  1  , N7    1  1  , N9    1  1 
4 4 4 4
For mid nodes:
N2 
1
2
   1
   1
   1
 
1   2  2   , N 4   2   1   2 , N 6   2   1   2 , N8  1   2  2  
2 2 2

For center node:
N5  1   2 1  2 
Further, the nodal unknowns at any generic point within FGM plate/shell can be expressed
in terms of shape function Ni. Hence, the displacement vector of node i in the mid-plane is
given by
NN
ui   Ni  , i  i (3.9.3)
i 1
where {u}i is the displacement vector within the element corresponding to node i, Ni is the
interpolation or shape functions expressed in terms of normalized natural co-ordinates (
and ), {δ}i is the nodal displacement vector and NN is the number of nodes in the
element. In order to circumvent the shear locking problem associated with the element, the
selective reduced integration technique has been employed in the formulation. For the
present nine-node isoparametric element, 2×2 Gauss integration is employed for shear
terms while 3x3 integration rule is employed for bending terms.
102
3.10 DISPLACEMENT FIELDS
3.10.1 Displacement field having constant variation of transverse
displacement
To describe the deformation profile of the plate/shell, a special form of displacement field
proposed by Reddy (Reddy 1984) is chosen, where the in-plane displacement fields (u and
v) are expanded as cubic functions of the thickness term (z), and the transverse displacement
(w) has been assumed to be constant through the thickness. Any other choice of kinematics
field would either not satisfy the shear stress-free boundary conditions or lead to a theory
that would involve more dependent nodal unknowns than those in the FSDT (Reddy 1984).
In addition, the higher order theory leads to the parabolic distribution of transverse shear
stresses and therefore the need of any shear correction co-efficient could be eliminated.
According to Reddy’s higher order shear deformation theory (Reddy 1984), the in-plane
displacement components (u and v) and transverse displacement (w) are expressed in terms
of corresponding displacements at the mid surface (uo, vo and wo) by the following
expression.
u ( x, y, z )  u0 ( x, y )  z x ( x, y )  z 2 x ( x, y )  z 3 x ( x, y )
v( x, y, z )  v0 ( x, y )  z y ( x, y )  z 2 y ( x, y )  z 3 y ( x, y ) (3.10.1)
w( x, y )  w0 ( x, y )
where u, v and w are the displacements of any generic point in the plate/shell geometry. The
parameters u0, v0 and w0 are the displacements of points which are lying on the mid-surface
(i.e., reference surface); and θx, θy are the bending rotations defined at the mid- surface about
the y and x axes respectively. ξx, ξy, x and y are higher order terms appears in Taylor’s
series expansion and solved by the condition of zero transverse shear stains (γxz (x, y, ± h/2) =
γyz (x, y, ± h/2)=0) at the top and bottom of the shell surface. Thus, incorporating the above
boundary condition in Eq. (3.10.1) leads to the expression for unknown higher order terms
(ξx, ξy, x and y). Finally, by substituting the values of unknown all the higher order terms
(ξx, ξy, x and y) in Eq. (3.10.1) and rearranging the terms that appears in the displacement
field (u and v), the following final expression may be obtained.
103
4z3  w 
u ( x, y, z )  u0  z x   
2  x 
3h  x 
4z3  w 
v( x, y, z )  v0  z y   
2  y  (3.10.2)
3h  y 
w( x, y )  w0
If the displacement field represented in Eq. (3.10.2) is implemented in the strain
part, the problem of C1 continuity requirement invited by higher order theory may arise
due to the presence of first order derivatives of transverse displacement. To circumvent the
problem, the derivatives of the transverse displacement are treated as separate field
w   w 
variables  x    x   and  y    y   . Hence, by incorporating the above substitution,
 x   y 
the final form of higher order theory possessing C0 continuity may be arrived in the
following manner.
4z3
u ( x, y, z )  u0  z x  x
3h 2
4z3
v( x, y, z )  v0  z y  2  y (3.10.3)
3h
w( x, y )  w0
Hence, the basic field variables interpreted in the present investigation with the
assumption of constant transverse displacement component are u0, v0, w0, θx, θy, γx and γy for
each node.
Further, the expression of strain vector {ε} can be correlated with displacement
vector {𝛿} by means of following relationship.
    B  (3.10.4)
where [B] is known as strain-displacement matrix having the order 15x7 and involves the
derivatives of shape function terms.
The nodal displacement vector {𝛿} corresponding to displacement field in Eq.
(3.10.3) may be written as    u0 v0 w0  x  y  x  y 

T
(3.10.5)
104
3.10.1.1 Strain-displacement relationship for plate
 u0  z  x  4 z  x 
3
 u
 x   
x x 3h x 2
   
 v   v0  z  y  4 z  y 
3
 y   y y 3h y 2 
   
 u v   u0  x 4 z 3  x v0  y 4 z 3  y  (3.10.6)
 m         z  2  z  
 y x   y x 3h x y y 3h 2 y 
 u w   w 4z2 
    0 x  2  x 
 z x   x h 
 v w   w0 4 z 2 
 z  y   y  2  y 
   y h 
Since the plane stress problem is considered (𝜎zz=0) in the analysis, the components of
mechanical strain vector may be represented as in Eq. (3.10.6).
For thermo-mechanical analysis of the plate, the total strain can be expressed in terms of
mechanical and thermal strain as
    m    t  (3.10.7)
Further, the mechanical strain vector in terms of total strain can be rewritten as
 m    H  p  
1
(3.10.8)
where the matrix [H]p1 contains thickness and other zero terms and given in Appendix B.
Since the effect of thermal field is to expand and contract the cross section, the expression
  z , T  
 

h  z, T
  
2
 
for thermal strain involves  t    0  dz (3.10.9)
h  
2 
0

0 
 
Hence total strain with the help of Eq. (3.10.7) and (3.10.8) can be described as
    H     t  (3.10.10)
3.10.1.2 Strain-displacement relationship for shell

The linear strain- displacement relationship according to Sander’s shell theory incorporating
all the three radii of curvature may be written as,
105
u w
x  
x Rx
v w
y  
y Ry
u v 2w
 xy    (3.10.11)
y x Rxy
u w C1u C1v
 xz    
z x Rx Rxy
v w C1v C1u
 yz    
z y Rx Rxy
where Rx, Ry represents the radii of curvature in the x and y directions respectively and Rxy is
the twist radii of curvature. C1 is the tracer that helps to reduce the approximation in to
Love’s shell theory and assumed as unity in the present analysis. The strain terms may be re-
arranged in the following manner by the combination of Eqs. (3.10.3) and (3.10.11).
 4z2  4z3 *
 x   xo  z 1   x
k  kx
 3h 2  3h 2
 4z2  4z3 *
 y   yo  z 1  2  k y  2 k y
 3h  3h
 4z2  4z3 *
 xy   xyo  z 1  2  k xy  2 k xy (3.10.12)
 3h  3h
 4z2  4 z 3 * 4 z 2 **
 yz   y  z 1  2  k yz  2 k yz  2 k yz
 3h  3h h
 4z2  4z3 4z2
 xz  x  z 1  2  k xz  2 k xz*  2 k xz**
 3h  3h h
where the different terms involved in Eq. (3.10.12) are defined in the following fashion.
 x y y x  x*  *y 
C1  C1 , C1  C1 , C1  C1 ,
 R R R R R Rxy 
kxz , k yz , kxz* , k *yz , kxz** , k yz**    * x xy y xy x

  x* 
  , x  x* , y  *y
y
C
 1R C1 
 y R xy 
 
 xo ,  yo ,  xyo    uxo  wRo , vyo  wRo , uyo  vxo  2Rwo 
 x y xy 
 
x , y    uz  wx  CR1u  CR1v , wy   y  CR1vo  CR1uo  (3.10.13)
 x xy x 
xy 
  x  y  x  y  x*  *y  x*  *y 
kx , k y , kxy , k , k , k    x , y , y  x , x , y , y  x 
*
x
*
y
*
xy
 
106
Further, the mechanical strain vector in terms of total strain can be expressed as
 m    H s   (3.10.14)
where the matrix [H]s contains thickness and zero terms.
3.10.2 Displacement field having quadratic variation of transverse

displacement
To predict the overall response of the plate in accurate sense, it is necessary to represent
the kinematics field that contains membrane and flexure terms in in-plane and transverse
displacement expression. To address the necessity, a kinematics field that captures normal
and transverse cross sectional deformation modes is considered in the present research
work. The assumed in-plane fields incorporate the cubic degree of thickness term, where in
the quadratic degree of thickness term is used for the transverse component. Hence, at any
general point P (x, y and z), the in-plane (u and v) and transverse displacement (w) field is
incorporated in the following manner (Kant and Swaminathan 2000).
u ( x, y, z)  uo ( x, y, z)  z x ( x, y, z)  z 2 x ( x, y, z)  z 3  x ( x, y, z)
v( x, y, z)  vo ( x, y, z)  z y ( x, y, z)  z 2 y ( x, y, z)  z 3  y ( x, y, z) (3.10.15)
w( x, y, z)  wo ( x, y, z)  z z ( x, y, z)  z 2 z ( x, y, z)
The terms u0, v0, x and y assume the similar meaning as in Eq. (3.10.1). The
polynomial functions related to the quadratic and cubic terms in the in-plane and transverse
field are known as Taylor’s higher order terms and considers the shear deformation modes
in the thickness direction. To solve for these higher order polynomial terms stress free
boundary condition are imposed at the top and bottom of the plate (i.e., ηxz= ηxz=0 at z=
±h/2). Hence the appropriate substitution of variables from the boundary condition leads to
the following expression for in-plane displacement components.
 4z3  z2  z 4 z 3 w0 z 3  z
u ( x, y, z)  uo   z  2   x   
 3h  2 x 3h 2 x 3 x
(3.10.16)
 4z3  z2  z 4 z 3 w0 z 3  z
v( x, y, z)  vo   z  2   y   
 3h  2 y 3h 2 y 3 y
The strain expression related to the above displacement field leads to the existence
of second order derivatives of transverse displacement in the expression because of the
presence of first order derivatives in displacement field. But, in most practical applications
C0 elements are generally preferred compared to C1 elements because of their easy
107
isoparametric formulation. Hence to develop C0 formulation, the out of plane derivatives
are replaced by the following suitable independent field variables.
 z w   w 
x   0;  x  0  0;  x  z  0;  y  z  0;  y  0  0; y  z  0 (3.10.17)
x x x y y y
The final form of displacement field (u, v and w) incorporating C0 variables becomes
 4z3  z2 4z3 z3
u ( x, y, z)  uo   z  2   x   x  2  x   x
 3h  2 3h 3
 4z3  z2 4z3 z3
v( x, y, z)  vo   z  2   y   y  2  y   y (3.10.18)
 3h  2 3h 3
w( x, y, z)  wo  z z  z 2 z
Further, the strain vector {ε} can be correlated with displacement vector {𝛿} by means of
following relationship.
    B  (3.10.19)
where [B] is the strain-displacement matrix having the order of 20x13 and involves the
derivatives of shape function terms.
The nodal displacement vector {𝛿} for the displacement field in Eq. (3.10.18) may be
written as,    u0 v0 w0  x  y  z  x  y  z  x  y  x  y  T (3.10.20)
3.10.2.1 Strain-displacement relationship for FGM plate
The linear strain-displacement relationship for the plate element incorporating the effect of
normal strain may be expressed as
u v w
 xx  ;  yy  ;  zz 
x y z
(3.10.21)
u v u w v w
 xy   ;  xz   ;  yz  
y x z x z y
Further incorporating expressions provided in Eq. (3.10.18) into Eq. (3.10.21), leads to the
following correlation between strain and displacement terms (Eq. (3.10.22)).
108
uo  4 z 3   z 2  x 4 z 3  x z 3  x
 xx   z  2  x   
x  3h  x 2 x 3h 2 x 3 x
vo  4 z 3   y z 2  y 4 z 3  y z 3  y
 yy   z  2    
y  3h  y 2 y 3h 2 y 3 y
 zz   z  2 z z
 uo vo  4 z 3   x  4 z 3   y z 2  x 
    z     z    
 y x  3h 2  y  3h 2  x 2 y 
 xy   (3.10.22)
 z  y  4 z  x  4 z  y  z  x  z  y 
2 3 3 3 3
 2 x 3h 2 y 3h 2 x 3 y 3 x 

 4z2  4z2 w  
 xz  1  2   x  z x  2  x  z 2 x  o  z z  z 2 z
 h  h x x x
 4z2  4z2 w  
 xy  1  2  y
  z   y  z 2 y  o  z z  z 2 z
y y y
y 2
 h  h
It is to be noted that Eq. (3.10.22) contains only first order derivatives of transverse
displacement component in the derived expression, and thus leading to efficient C0
formulation.
The strains associated with Eq. (3.10.22) are related to the generalized strains by means of
the following expression.
 m   H  p  
2
(3.10.23)
where  m   xx  yy  zz  xy  xz  yz 
T
T
 uo vo  w   w   u v        
 , , z ,  o   y  ,  o  x  ,  o  o  ,  x  ,  y  ,z , 
 x y  y   x   y x   x   y  
              
 z   y  ,  z   x  ,  x  y  ,  x  ,  y  ,   y  z   y  y  , 
 y   x   y x   x   y   y 
    
    z     ,   x   y  ,   x   x   x  , 
 x x x     
  y x   x x x 
x
 
  y  y  y    x  x  x  y  y  y  
   ,       
 y y y   y y y x x x  
109
The components of matrix [H]p2 of order 6x20 contains the linear, quadratic and
cubic function of z terms and derived in Appendix B.
3.10.2.2 Strain-displacement relationship for FGM shell

The linear strain-displacement relation for the assumed shell geometry incorporating radius
of curvature in x and y directions in addition to twist curvature in x-y plane may be
expressed as
u w
 xx  
x Rx
v w
 yy  
y Ry
w
 zz 
z
u v 2 w (3.10.24)
 xy   
y x Rxy
u w u v
 xz    C1  C1
z x Rx Rxy
v w v u
 yz    C1  C1
z y Ry Rxy
Further, incorporating Eq. (3.10.18) into Eq. (3.10.24) leads to the following strain field
expression.
z2 I 4z3 z3
 xx   x 0  z x   x  z 2 x II  2  x III   x IV
2 3h 3
2 3
z 4z z3
 yy   y 0  z y   y I  z 2 y II  2  y III   y IV
2 3h 3
 zz   z 0  2 z z
z 2 II 4 z 3 IV z 3 V
 xy   xy 0  z xy  2 z   xy  2 z  xy  2  xy   xy
I
xy
2 III (3.10.25)
2 3h 3
2 2 3 3
4z z 4z z
 xz  x  z xz  2  xzI   xzII  z 2 xzIII  2  xzIV   Vxz
h 2 3h 3
2 2 3
4z z 4z z3
 yz   y  z yz  2  yzI   yzII  z 2 yzIII  2  yzIV   Vyz
h 2 3h 3
110
where
 u w   v w   u v 2w  
 x0 ,  y 0 ,  z 0 ,  xy 0    0
 0  ,  0  0  ,  z ,  0  0  0 
   y x R 
 x Rx   y Ry   
xy  
 w0 v   w u  
x , x     x 
u v
 C1 0  C1 0  ,   y  0  C1 0  C1 0  
 x Rx Rxy   y Rx Rxy  
 , 
x y ,  xy ,  x I ,  x II ,  x III ,  x IV ,  y I ,  y II ,  y III ,  y IV ,  xyI ,  xyII ,  xyIII ,  xyIV ,  Vxy 
      y      y                 y     
 x  z  ,   z , x 
 ,
x
, ,
z x
 x , x
,  ,   , 
z
 x Rx   y R y    y x   x   x 
R x x   x   y   y 
R
 
  y  y    y    z    x  y    z    x  y  x  y    x  y  
 y  y  ,  y  ,  R  ,  y  x  ,  R  ,  y  x  y  x  ,  y  x  
     xy     xy     
 xz ,  yz ,  xzI ,  xzII ,  xzIII ,  xzIV ,  Vxz ,  yzI ,  yzII ,  yzIII ,  yzIV ,  Vyz 
   y  z           
 C1 x  C1    x  ,  C1 y  C1 x  z   y  ,  x   x  ,  C1 x  C1 y  , 
 Rx Rxy x   Ry Rxy y   R Rxy  
    x
 
     y  y    x y   y  
  x  z  ,  C1 x  C1  C1 x  C1  ,  C1  C1  ,  y   y  ,  C1  C1 x  , 

x   Rx    
 Rxy Rx Rxy   Rx Rxy   Ry Rxy 
 
   z  ,  C  y  C  x  C  y  C  x  ,  C  y  C  x  
 y y   1 R 1
Rxy
1
Ry
1   1
Rxy   Ry
1 
Rxy  
  y 
The strain vector associated with the displacement field represented in Eq. (3.10.25) may
be represented in terms of generalized strains as
 m   H s   2
(3.10.26)
where  m   xx  yy  zz  xy  xz  yz  and

T
T
 x ,  y ,  xy ,  x ,  x ,  x ,  x ,  y ,  y ,  y ,  y ,  xy ,  xy ,  xy ,  xy ,  xy , 
 I II III IV I II III IV I II III IV V

    

 xz ,  yz ,  I
xz ,  II
xz ,  III
xz ,  IV
xz ,  V
xz ,  I
yz ,  II
yz ,  III
yz ,  IV
yz ,  V
yz 

3.10.3 Governing equation for Static analysis

The total potential energy of the structural system under the action of externally applied
transverse load may be represented as
 p  U s  We (3.10.27)
where Us is the strain energy of the system and We is the potential of the externally applied
load.
111
The expression for strain energy may be given as
1
Us       dx dy dz
T
(3.10.28)
2 v
By utilizing the relationship given in Eq. (3.9.1) or Eq. (3.9.2) depending upon the
assumption of transverse displacement component, the strain energy in the system gets
modified into following form.
1
   D  dx dy
T
2 
Us   (3.10.29)
where  D    H  Q H  dz
T
(3.10.30)
The matrix [H] appears in Eq. (3.10.30) represents the general form and has to be replaced
by the appropriate expression from Eq. (3.10.8), Eq. (3.10.14), Eq. (3.10.23), and Eq.
(3.10.26) depending upon the displacement field and type of geometry (plate/shell).
Further, the potential energy due to the externally applied load of intensity q can be
considered in the following form. We   w p dx dy (3.10.31)
where  p  0 0 q 0 0 0 0 0 0 0 0 0 0
Combining either Eq. (3.10.3) or Eq. (3.10.18) depending upon the displacement field,
with Eq. (3.10.27), Eq. (3.10.29) and Eq. (3.10.31) may leads to following expression of
total potential energy as
1
d  B  D Bd  dx dy   d   N w   p dx dy
T T

T
p 
T
(3.10.32)
2
where [Nw]T represents the shape function like matrix with non-zero terms associated with
the corresponding transverse nodal displacement. According to the principle of minimum
potential energy, the first variation of  must be zero for equilibrium condition. Thus
taking the first variation of potential energy, Eq. (3.10.32) yields into following form.
  p   d 
T
   B  D Bd dx dy    N   p dx dy
T w T
 (3.10.33)
Since the expression present in the above equation should hold good for any variation of
{δd}, the quantity present in Eq. (3.10.33) must be equated to zero.
  B  D Bd dx dy    N   p dx dy  0
T w T
(3.10.34)
Further incorporating the terms associated with stiffness matrix and nodal load vector at
element level, the final governing equation for bending analysis may be expressed as
112
 K d  p (3.10.35)
The stiffness matrix [K] and nodal load vector {p} formed at element level are further
assembled to get stiffness matrix and load vector at global level. This can be achieved by
taking the contribution of all the plate/shell elements in the formulation. The solution of
linear simultaneous equations consumes significantly more computational time in FE
analysis. Hence by considering the symmetric and banded nature of the global stiffness
matrix a special algorithm known as Gauss elimination technique is used to decompose the
equations. The nodal displacements may be obtained for the static problem by
incorporating appropriate boundary conditions. Finally, the in-plane and shear stresses are
computed with the substitution of stress-strain relationship established in Eq. (3.9.1) or Eq.
(3.9.2) depending upon the type of problem.
3.11 GOVERNING EQUATION FOR DYNAMIC ANALYSIS

3.11.1 Free vibration analysis
The acceleration at any point within the element may be expressed in terms of the mid
surface displacement parameters (u0, v0 and w0) as
u0 
2 

 f   2 f   2 v0    2  F  f 
..
(3.11.1)
t w 
 0
where the vector {f} represents the nodal unknowns which is of order 7x1 and contains the
terms of Eq. (3.10.3) and of the order 13x1 having the terms of Eq. (3.10.18).
Again the matrix {f} is decoupled into matrix [C] that contains the shape functions (Ni) and
global displacement vector {𝛿}.
 f   C   (3.11.2)
By utilizing the Eq. (3.11.1) and Eq. (3.11.2), the mass matrix of an element may be
expressed as,
 m   C   LC  dA
T
(3.11.3)
A
where the expression of matrix [L]can be written as
 L     F   F  dz
T
(3.11.4)
z
113
where ρ is the density of the composite material estimated from Eq. (3.4.2). Hence the
governing equation for free vibration analysis becomes,
 K     M  X   0
2
(3.11.5)
The right hand side zero in Eq. (3.11.5) represents the problem of free vibration analysis.
The eigen value algorithm is utilized to extract the mode shapes of the plate/shell problem. If
thermal effect is considered in the analysis the expression for governing equation becomes
  K    M  X   0
' 2
(3.11.6)
in which  K    K   T  K G
'
where [M], [K’], [K], [K]G, and ∆T are mass matrix, reduced stiffness matrix due to thermal
load, linear stiffness matrix, geometric stiffness matrix and critical buckling temperature
parameter, respectively.
3.11.2. Forced vibration analysis

For the problem of forced vibration Eq. (3.11.6) is modified to incorporate damping matrix
[C] at left and the force vector {q} at the right hand side. Hence the governing equation for
forced vibration analysis may be expressed as
.. .
[M ]U  [C ]U  [ K ]U  {q} (3.11.7)
Where, [M] and [K] implies the global mass matrix and stiffness matrix, respectively. The
matrix [C] is the Rayleigh damping matrix and considered as below.
[C]   [M ]   [ K ] (3.11.8)
In Eq. (3.11.7), α and β are constants to be determined from two given damping ratios
corresponding to two unequal frequencies of vibration. The force term {q} appearing in Eq.
(3.11.6) is the dynamic pressure applied on the top of the shell and represented as,
q  x, y, t   q0 F (t ) (3.11.9)
where q0 is the maximum amplitude and F(t) is a dynamic load shape function of time
domain. In the present analysis F(t) is taken as unity for the case of suddenly applied load.
The extension of the linear acceleration method known as Newmark integration method
(Bathe 1994) is utilized to obtain the transient response of the structural system. A step by
step procedure for the problem of forced vibration response is summarized below.
114
 For the problem under consideration the stiffness matrix [K], mass matrix [M] and
damping matrix [C] are computer at global level as an initial step.
The magnitude of displacement U  , velocity  U  and acceleration  U  at time t=0

. ..

   
are initialized in the problem.

 The appropriate time step ∆t is chosen, and parameters α and β have to be determined
from the damping ratios that correspond to two unequal natural frequencies that
obtained from free vibration response.
 The various co-efficients are determined from the expression given below.
1  1 1 
ao  ; a  ; a  ; a   1; a   1;
t 2
1
t 2
t 3
2
4

t    (3.11.10)
a5    2  ; a6  t (1   ); a7  t
2  
 Evaluate the effective stiffness matrix [K’] as  K '  [ K ]  a0  M   a1 C 

(3.11.11)
 The above generated effective stiffness matrix [K’] is triangularized. Then, the
effective loads [R’] are estimated at time t+∆t.
 From the effective stiffness matrix [K’] and load matrix [R’] generated from above
step, the displacement is solved for time t+∆t and subsequently the velocity and
acceleration can be estimated at time interval t+∆t.
3.12 GOVERNING EQUATION FOR BUCKLING ANALYSIS

In order to derive the geometric stiffness matrix the geometric strain vector may be
expressed as (Cook et al., 2004)
1
 G   AG  
2 
(3.12.1)
 w u v 
 0 0 0
 x x x 
 w u v   w w u u v v 
T
where  AG    0 0 0  and     
 y y y   x y x y x y 
 
 w w u u v v 
 y x y x y x 
Further, using strain-displacement relationship given in Eq. (3.10.4) or Eq. (3.10.19), the
geometric stiffness matrix [kg] may be represented as
115
kg  
  B G  B dx dy
T
  (3.12.2)

where G   H G   S  H G dz
T i
(3.12.3)
The expression of [HG] appear in Eq. (3.12.3) is derived in Appendix A. The stress matrix
 S i  can be formed in terms of in- plane and shear stresses as mentioned below.
 
 x  xy 0 0 0 0 
 
 xy  y 0 0 0 0 
 0 0  x  xy 0 0 
S i   
   0 0  xy  y 0 0  (3.12.4)
 
 0 0 0 0  x  xy 
 
 0 0 0 0  xy  y 
Hence the final governing equation for linear buckling analysis can be written in the
following form.
[ K ]   [ K ]G    0 (3.12.5)
where  is the critical buckling load, {δ} is the nodal displacement vector and [K], [K]G are
the linear and geometric stiffness matrices, respectively, for the analysis. If thermal effect
is considered for buckling problem, thermal stresses developed at different gauss points of
the element due to assumed temperature rise are calculated initially by solving the
corresponding bending problem. The calculated thermal stresses are then used to generate
stress matrix [Si] of thermal buckling problem that exists in the geometric stiffness matrix
[K]G.
3.13 GEOMETRICALLY NON-LINEAR ANALYSIS

3.13.1 Green-Lagrange strain displacement relation
To incorporate the non linear relationship between strain and displacement, the expression
of Green’s strain vector (Cook 2004) may be represented as
116
 u 1  u  2 1  v  2 1  w  2 
          
 x 2  x  2  x  2  x  
 2 2 2 
 x   v  1  u   1  v   1  w  
   y 2  y  2  y  2  y  
 y   
   u v u u v v w w 
   xy  
      .  .  .  (3.13.1)
   y x x y x y x y 
 xz   u w u u v v w w 
 yz     .  .  . 
   z x x z x z x z 
 v w u u v v w w 
 z  y  y . z  y . z  y . z 
 
 
By the accomplishment of von Kármán large deflection assumptions, which implies that
derivatives of u and v with respect to x, y and z are small and w is independent of z allows
Green’s strain to be rewritten in terms of the mid-plane deformations of Eq. (3.10.3) as,
 u 1  w  2 
    
 x 2  x  
 2 
 x   v  1  w  
   y 2  y  
 y   
   u v w w 
 xy      .  (3.13.2)
   y x x y 

   u w
xz 
 yz    
   z x 
 v w 
 z  y 
 
 
Combination of Eq. (3.10.3) and Eq. (3.13.2) leads to following form of expression for the
strain field.
T
  uo vo vo uo     w w      
  x  y x y x  y 0 0  x  y x y x  y   
  x y x y x y y x x y x y y x  
  
  2
  (3.13.3)
     
2
  1  w  1 w w w 
  2  x  2  y  x y
. 0 0 0 0 0 0 0 0 0 0 0 0

   
or, in compact form Eq. (3.13.3) may be represented as
    L   NL (3.13.4)
where  L and  NL are the linear and non-linear strain terms, respectively. In order to
develop the FE equations, the virtual work for a continuum are expressed in terms of total
117
Lagrangian co-ordinate system under the assumption of small strains and conservative
loading as
 d    dV   du q dV   du P dS

T T T
(3.13.5)
V V S
where V is the undeformed volume, S is the portion of the surface on which forces are
defined, {ζ} the stress vector, {dε} is the virtual Green’s strain vector due to the virtual
displacements {du}, {q} represent the body forces per unit mass and {P}the surface
traction acting over an undeformed surface dS. The virtual work given in Eq. (3.13.5) may
now be written entirely in terms of mid-plane displacement quantities as
 d    dA   du q dA   du P dS

T T T
(3.13.6)
A A S
Before proceeding with the discretisation of the virtual work presented in Eq. (3.74), it
 
becomes necessary to consider further the variation of strain d  due to variation of
   
displacements du . Generally, d  is given as the sum of the variation of the linear and
non-linear generalized strains as
d   d   d 

L NL
(3.13.7)
The displacement field within an element at any point may be given as a function of n
discrete nodal displacements.
n
ui   Ni  ,   (3.13.8)
i 1
where n represents the total number of nodes in the element, Ni defines the shape functions
associated associate with node i in terms of normalized coordinates  and , and   is
the generalized displacement vector and can be represented with the help of Eq. (3.10.5)
or Eq. (3.10.20).
With the help of Eq. (3.13.8), the linear strain  L can be represented as
n
 L    BL  i (3.13.9)
i 1
where  B L denotes the linear strain-displacement part giving the linear  L of in-plane,
bending and shear strains, which is linearly dependent on the displacement vector   . The
expression for displacement gradient with respect to the lateral displacements w0 may be
denoted as
118
T
 w w 
     (3.13.10)
 x y 
Hence with the guidance of Eq. (3.13.10), the non-linear strain vector  NL can be
formulated as,
n
 NL    BNL  i (3.13.11)
i 1
where the non-linear strain-displacement matrix  B NL may be expressed in terms of
1
displacement gradient given in Eq. (3.13.10) as  B NL   A  (3.13.12)
2
Further the expression for matrix [A] can be expressed as,
 w 
 0 
 x 
 w 
 A   0  (3.13.13)
 y 
 w w 
 
 y x 
Combination of Eq. (3.13.8), Eq. (3.13.10) and Eq. (3.13.13) directed to the following
expressions for [A] and {θ}.
 n  N i  
   x wi 0 
 i 1   
 n
 N i  
0
 A  

i 1  y
wi  (3.13.14)
 
 n  N  
  i
n
 N i 
 i 1  y
wi   x
wi



 i 1
 
 
n
    G i  i (3.13.15)
i 1
 Ni 
0 0 x
0 0 0 0
where G i    (3.13.16)
0 0 Ni
0 0 0 0
 y 
and the initial stress matrix [S] involves the following stress terms.
 x  xy 
S     (3.13.17)
 xy  y 
119
With the help of Eq. (3.13.9) and Eq. (3.13.11), the total strain matrix can be written as
sum of linear and non-linear strain matrices.
    BL   BNL    

1
(3.13.18)
 2 
For every element the linear and nonlinear strain-displacement matrices, respectively,
 B L and  B NL are assembled in terms of nodal sub matrices as,
 BL   BL(1) ,  BL(2) ,........... BL(n)  (3.13.19)
 BNL   BNL(1) ,  BNL(2) ,........... BNL( n)  (3.13.20)
Further, the virtual work Eq. (3.13.6) is discretisized for an element, by substituting strain
and displacement variations of Eq. (3.13.8) and Eq. (3.13.18) as
  1   
d      B L   B NL      dA   R   0
T
(3.13.21)
 A  2   
where the equivalent nodal load vector R due to body forces and tractions may be written
as
R    N  q dA    N  P dS
T T
(3.13.22)
A S
Since the nodal virtual displacements d  are arbitrary, the element non-linear equilibrium
becomes,
      B    dA  R  0
T
(3.13.23)
A
where the total strain matrix is nothing but  B   BL   B NL . The load term R may also
contain nodal point loads. The Eq. (3.13.23) can have the dual role of representing either
the element, or in an assembled form the total equilibrium equation. It represents a
nonlinear equation in   , since  B  and   denotes linear and quadratic functions of   ,
respectively.
3.13.2 Solution to non-linear equilibrium equations

The obtained solution algorithm for the assembled nonlinear equilibrium equations are
based on the Newton-Raphson (N-R) iterative scheme which consists of solution of a
series of linear equations. The tangent stiffness matrix  KT  may be formed at each
iteration level as a function of calculated displacement in corresponding step. If an initial
120
estimate  i for the total displacements gives residual (unbalanced forces)   i  0
than an enhanced value  i 1 may be obtained by equating to zero the linearised Taylor’s
series expansion of   i 1 in the neighborhood of  i as
  i 1    i   KT  i  0 (3.13.24)
where  KT  is termed as the assembled tangent stiffness matrix evaluated at  i and
represented by
   i 
 KT     (3.13.25)
  
The expression in Eq. (3.13.24) is the linear incremental equilibrium equation which gives
the linearised approximation to the correlation between the residual forces and incremental
displacements  i at a point  i on the non linear equilibrium path. The improved
solution may then be identified as,

 i 1   i   i (3.13.26)
The expressions appear in Eqs. (3.13.24) and (3.13.26) represent the Newton-Raphson
solution to the nonlinear Eq. (3.13.23). The term   i and  i denotes the measures of
the convergence of the solution. To assist the numerical stability of the process the load R
is applied in a series of steps from 1 to k, a typical step having a total applied load Rk
k
where R k   R j (3.13.27)
j 1
and equilibrium is established for every Rk value. The total displacements  k are defined
within each load step k; number of iterations i take place until convergence is achieved;
consequently the total incremental displacement  n is summed up as follows.
k
  n    n (3.13.28)
j 1
Hence the final equilibrium equation and its incremental form is specified as
 KS    R (3.13.29)
 KT d   dR (3.13.30)
where
1 1
 KS     B  D BLdV    B  D BNL dV    B   D B L dV    B   D  B L dV (3.13.31)
T T T T
V
L
2 V
L
V
NL
2 V
NL
121
For the nodal unknowns corresponding to all the nodes of elements, the expression for
tangent stiffness matrix  KT  may be assembled as
  1  1  1 
   ................  
 1 2 m

  2  2  2 
   ................  
 1 2 m 
 KT    . .  (3.13.32)
 
 . . 
 . . 
 
  m  m ................  m 
 1  2  m 

Solving each term appear in the expression of  KT  , the final equation takes the following
form.
 KT     B  D B dA   G   S i G  dA
T T
(3.13.33)
A A
Substitution of Eqs. (3.13.16), (3.13.17) and (3.13.18) in to Eq. (3.13.33) gives the final
form of  KT  as
 KT    K0    K L    K  (3.13.34)
in which the constant linear elastic stiffness matrix  K 0  can be expressed as
 K0     B  D BL dA
T
(3.13.35)
L
A
The initial displacement matrix  K 0  that quadratically dependent upon displacements  
may be represented as
 K L     B  D  B  NL dA    B NL  D  B L dA    B NL  D  B NL dA

T T T
(3.13.36)
L
A A A
and finally the initial stress matrix may be formed as
 K    G   S i G  dA
T
(3.13.37)
A
122
At each load increment, convergence is checked using total residual norm criteria as
  
T 2
100   1 (3.13.38)
1
 R R
T 2
where γ1 is the convergence tolerance (%).
3.14 FORMULATION FOR POST BUCKLING ANALYSIS

3.14.1 Green-Lagrange strain-displacement relations
Considering the contribution of all non-linear terms in Green-Lagrange equations,
the strain-displacement relations for a general problem of FGM can be expressed as
 u 1  u  2  v  2  w  2  
           
 x 2  x   x   x   
 
 v 1  u   v   w  
2 2 2

           
 xx   y 2  y   y   y   
   

 yy   w 1  u   v   w  
2 2 2

 zz             
       z 2  z   z   z   
 yz   v w  u u   v v   w w   
      .    .    .   
 xz   z y  y z   y z   y z   
 xy   
 u  w   u . u    v . v    w . w  
 z x  x z   x z   x z  
 
 u v  u u   v v   w w   
    .    .    .   
 y x  x y   x y   x y    (3.14.1)
The total strain   terms given in Eq. (3.14.1) can be separated to represent the linear
 l and non-linear  nl terms in the following manner.
    l   nl (3.14.2)
The separation of linear strain terms present in Eq. (3.14.2) leaves the following terms in
linear strain vector  l .
T
 u v w  u v   u w   v w  
 l           (3.14.3)
 x y z  y x   z x   z y  
123
Further, the quantities present in Eq. (3.14.3) are expressed by means of mid-plane strain
terms with the help of Eq. (3.10.18) as
 lxx  10  zk11  z 2k12  z 3k13 ; lyy   20  zk21  z 2k22  z 3k23 ; lzz   30 ;
(3.14.4)
    zk  z k ;    zk  z k ;    zk  z k  z k .
l
yz 0
4
1
4
2 2
4 l
xz 0
5
1
5
2 2
5 l
xy 0
6
1
6
2 2
6
3 3
6
The superscripts ‘0’, ‘1’, ‘2-3’ present in Eq. (3.14.4) are membrane, curvature and higher
order terms, respectively. Separation of co-efficient and thickness terms (linear, quadratic
and cubic) leads to the following relation between linear and mid-plane strain terms.
 l   H l  l (3.14.5)
Mid-plane linear strain vector   present in Eq. (3.14.5) contains the nodal unknowns and
l
derivatives of nodal unknowns and populated in the following manner.
      lzz  lyz  lxz  lxy klxx klyy klyz klxz klxy klxx klyy klyz klxz klxy klxx klyy klxy 
T
xx yy
l0 l0 0 0 0 0 1 1 1 1 1 2 2 2 2 2 3 3 3
l
T
 uo vo  w   w   u v       y  
 , , z ,  o   y  ,  o  x  ,  o  o  ,  x  ,   ,z , 
 x y  y   x   y x   x   y  
              
 z   y  ,  z   x  ,  x  y  ,  x  ,  y  ,   y  z   y  y  , 
 y   x   y x   x   y   y 
(3.14.6)
 
    z     ,   x   y  ,   x   x   x  , 
 x x x     
  y x   x x x 
x
 
  y  y  y    x  x  x  y  y  y  
   ,       
 y y y   y y y x x x  
The components of matrix  H l contain the linear, quadratic and cubic function of z terms.
Further, the linear mid-plane strain vector   can be correlated with displacement vector
l
  by means of linear strain-displacement matrix.

    B  
l l
(3.14.7)
where  B l is a matrix of differential operator that contains shape functions and   is the
displacement vector and formed by the contribution of all the nodes present in an element.
The non-linear strain vector   nl
can be expressed with the help of Eq. (3.14.1) as
124
 u  2  v  2  w  2  
         
 x   x   x   
 
 u   v   w  
2 2 2


          
 y   y   y   
 
 u    v    w  
2 2 2

        
  z   z   z  
1
 nl 
2
 u u   v v   v v  
2  .    .    .  
  y z   y z   y z  
 
2  u . u    v . v    v . v   
  x z   x z   x z   
 
  u u   v v   v v  
2  .    .    .  
  x y   x y   x y   (3.14.8)
Upon substitution of Eq. (3.10.18) into Eq. (3.14.8) leads to the following membrane,
curvature and higher order terms to be present in non linear strain vector   nl
.
 nlxx 
1 xx
2
 nl0  zknlxx1  z 2 knlxx2  z 3knlxx3  z 4 knlxx4  z 5knlxx5  z 6knlxx6 ; 
1

 nlyy   nlyy0  zknlyy1  z 2 knlyy2  z 3knlyy3  z 4 knlyy4  z 5knlyy5  z 6knlyy6 ;
2

1

 nlzz   nlzz0  zknlzz1  z 2 knlzz2  z 3knlzz3  z 4knlzz4 ;
2
 (3.14.9)
 nl   nlyz0  zknlyz1  z 2 knlyz2  z 3knlyz3  z 4 knlyz4  z 5knlyz5 ;

yz
 nlxz   nlxz  zknlxz  z 2 knlxz  z 3knlxz  z 4knlxz  z 5knlxz ;

0 1 2 3 4 5
    zk  z k  z k  z k  z k  z 6knlxy .
xy
nl
xy
nl0
xy
nl1
2 xy
nl2
3 xy
nl3
4 xy
nl4
5 xy
nl5 6
Similar to linear strain vector  l , the non-linear strain vector  nl can be related to non-
linear mid-plane strain part   nl

by means of following equation.
 nl   H nl  nl (3.14.10)
 nlxx0  nlyy0  nlzz0  nlyz0  nlxz0  nlxy0 knlxx1 knlyy1 knlzz1 knlyz1 knlxz1 knlxy1 knlxx2 knlyy2 

 

where     knlzz2 knlyz2 knlxz2 knlxy2 knlxx3 knlyy3 knlzz3 knlyz3 knlxz3 knlxy3 
 xx yy zz yz xz xy xx yy yz xz xy xx yy xy 
nl

knl4 knl4 knl4 knl4 knl4 knl4 knl5 k nl5 k nl5 knl5 knl5 knl6 knl6 knl6  
125
where  H nl is a matrix containing terms which are function of thickness co-ordinate and
not shown here for the sake of large matrix size. The detailed expressions of different
terms present in  nl are given in Appendix C.
Further, the non linear strain part  nl can be rewritten in the simplest form as
1
 nl   A  (3.14.11)
2
The expression of matrix [A] contains displacement functions and the various terms appear
in the matrix [A] are given in Appendix D. Further, the matrix   can be expressed as
      
T
 uo uo vo vo wo wo
 , , , , , , x , y ,  z , x , x , y , y , z , z ,
 x y x y x y x y x y x y 
           
    x ,  y ,  z , x , x , y , y , z , z ,  x ,  y , x , x , y , y , 
 x y x y x y x y x y 
    y  y 
 x , y , x , x , , 
 x y x y 
The matrix   present in Eq. (3.14.11) can further be decomposed as

   G d (3.14.12)
 
Where G is a matrix contains differential operators of shape functions and   is the
displacement vector.
To derive the geometric stiffness matrix  K G , following relationship is established
among geometric strain vector  G and matrix  AG .

1
 G   AG   (3.14.13)
2
T
 u u u v v v w w w 
where     
 x y z x y z x y z 
Further, the matrix   and strain vector   can be related as
    H G   (3.14.14)
The terms of  H G matrix contain thickness-coordinate terms and the detailed
expression is given in Appendix A.
126
By combining Eq. (3.14.13) and Eq. (3.14.14) the expression for geometric stiffness
matrix  K G can be arrived and the expression is given below.
K     B G  B  dx dy
T
G
(3.14.15)
where G     H G  S   H G dz and the stress matrix [S]can be expressed in terms of in-

T
i
 xx  xy  zx 0 0 0 0 0 0
  yy  yz 0 0 0 0 0 0
 xy 
  zx  yz  zz 0 0 0 0 0 0
 
 0 0 0  xx
 xy
 zx
0 0 0 
 
plane and shear stresses as S   0 0 0  xy  yy  yz 0 0 0
 
i
0 0 0  zx  yz  zz 0 0 0
0 0 0 0 0 0  xx  xy  zx 
 
0 0 0 0 0 0  xy  yy  yz 
0 0 0 0 0 0  zx  yz  zz 

3.14.2. Strain energy of plate

The strain energy of the post buckled FGM plate can be expressed as
1
Us     Q   dx dy dz
T
(3.14.16)
2 v
Upon substitution of Eq. (3.9.2) and Eq. (3.14.2) into Eq. (3.14.16), the expression for
strain energy assumes the form below.
1  T
 nl Q nl  dx dy dz (3.14.17)
1 T 1 T 1 T
Us      Q        Q        Q    
2  v l l
2 l nl
2 nl l
4 
With the help of Eq. (3.14.5) and Eq. (3.14.10), the modified form of Eq. (3.14.17) can be
represented as
1  T 1 T 1 T 1 T 
Us    l  D1  l   l  D2  nl   nl  D 3  l   nl  D 4  nl  dxdy (3.14.18)
2 A 2 2 4 
where  D  is known as rigidity matrix for the particular layer depends of material
i
properties and height of that layer from neutral plane.

h h
2 2
 D    H  Q  H  dz; D     H  Q  H 
T T
1 l 2 nl
dz;
l l
h h
2 2
h h
2 2
 D    H  Q  H  dz; D     H  Q  H 
T T
3 l 4 nl
dz;
nl nl
h h
2 2
127
Combining Eqs. (3.9.3), (3.14.7) and (3.14.18), the expression of strain energy is possible
in the following way.
 T 1 
 i  B   D 1  B l  i   i  B   D 2  B    nl  i
T T T
1   2 
U s   
l l
 dxdy
2 A 1 1
  i  B     D 3  B l  i   i  B    D 3  B   nl  i  (3.14.19)
T T T T
 2 nl 4 nl 

3.14.3 Derivation of the governing equation

The governing equation for the post buckling of FGM plate under the action of in-plane
loads can be obtained by minimizing the total potential energy of the system.

0 (3.14.20)

where   U s  W
where W is the work done by the in-plane loads. With the help of Eqs. (3.14.16) - (3.14.20)
one can arrive the final governing equation for nonlinear buckling problem.
 1 1 
 K    K nl1   K nl2   K nl3      K G    0 (3.14.21)
 2 2 
where K  is the global stiffness matrix,  K nl ,  K nl and  K nl are the non linear
1 2 3
stiffness matrices which are defined as following.
 K     B   D 1  B  dA; K nl    B   D 2  AG  dA

T T
1
A A
 K nl   G   A  D 3  B  dA;  K nl   G   A  D 4  AG  dA

T T T T
2 3
A A
The post buckling strength of FGSP can be computed from Eq. (3.14.21) using a
linear/nonlinear eigen value analysis by means of direct iterative procedure. The following
procedure is employed to get the desired final solution of post buckling analysis.
i. The stiffness and geometric matrices are computed at element level using the
nonlinear FEM.
ii. The so formed stiffness and geometric matrices are assembled to get global
stiffness and geometric matrices.
iii. Once the required matrices are formed at global domain, Eq. (3.14.21) has been
solved as linear eigen value problem considering all the non linear stiffness
matrices to be zero.
128
iv. Using eigen value analysis, eigen vectors are extracted for corresponding eigen
values and eigen vector is normalized with its maximum value.
v. Further, normalized eigen vector is scaled up to the value of desired amplitude i.e.,
for instance wmax/h=0.2 (where Wmax is the maximum deflection and h is the
thickness of the plate) means the normalized eigen vector is scaled to 0.2 times the
total thickness.
vi. Having obtained the scaled eigen vector the non linear stiffness matrices are
formed at global level.
vii. The above procedure continues until two successive eigen values are converged to
a specified tolerance value (0.1%).
viii. The step 3 is repeated with the updated nonlinear stiffness matrices to obtain the
new eigenvalue and eigenvector and amplitude ratio is increased and the procedure
is repeated.
3.15 BOUNDARY CONDITIONS

Since the displacement based formulation is considered in the present study, it is sufficient
to apply kinematic boundary conditions to solve the governing equation. To implement the
boundary part, the diagonal terms of the global stiffness matrix [K] corresponding to the
restrained degrees of freedom are multiplied with a very high number (i.e., 1015). Different
combinations of boundary conditions adopted in the study are given below. However, the
formulation and developed coding does not pose any limitation on different restrained
degrees of freedom. A pictographic view of the FGM plate is depicted in Figure 3.13.
3.15.1 Boundary conditions considering constant transverse displacement

field
By considering the displacement field given in Eq. (3.10.3), following types of boundary
conditions are used in the analysis.
a) Simply supported (SSSS):
v = w = θy = γy = 0 at x= 0, a
u = w = θx= γx =0 at y= 0, b
b) Clamped (CCCC):
u = v = w = θx= θy= γy = γx =0 at x=0, a and y= 0, b
129
y
y =b
x=0 x=a
x
y=0
Figure 3.13 Skew plate co-ordinate system
c) Simply supported-clamped (SCSC):
v = w = θy = γy = 0 at x= 0, a
u = v = w = θx= θy= γy = γx =0 at y= 0, b
d) Clamped-free (CFCF):
u = v = w = θx= θy= γy = γx =0 at x=0, a
u = v = w = θx= θy= γy = γx ≠0 at y=0, b
3.15.2 Boundary conditions considering quadratic variation of transverse

displacement field
By considering the displacement field given in Eq. (3.10.18), following types of boundary
conditions are used in the analysis.
a) Simply supported (SSSS):
v0 = w0 = θy = ∝x =  z = βy = x = 0, at x = 0 and a.
u0 = w0 = θx = ∝y =  z = βx = y = 0, at y = 0 and b.
b) Clamped (CCCC):
u0 = v0 = w0 = θx = θy =  z = ∝x = ∝y =  z = βx = βy = x = y = 0, at x = 0, a and
y = 0, b.
c) Simply supported and clamped (SSCC):
v0 = w0 = θy = ∝x =  z = βy = x = 0, at x = 0 and a.
u0 = v0 = w0 = θx = θy =  z = ∝x = ∝y =  z = βx = βy = x = y = 0, at y = 0, b.
130
d) Clamped-free (CFCF):
u0 = v0 = w0 = θx = θy =  z = ∝x = ∝y =  z = βx = βy = x = y = 0, at x = 0, and y =
0.
u0 =v0= w0= θx= θy=  z =∝x=∝y=  z= βx= βy = x =y≠ 0, at x = a and y = b.
e) Simply supported-clamped (SCSC):

v0 = w0 = θy = ∝x =  z = βy = x = 0, at x = 0 and y = 0.
u0 = v0 = w0 = θx = θy =  z = ∝x = ∝y =  z = βx = βy = x = y = 0, at x = a, y = b.
3.16 COMPUTER IMPLEMENTATION

3.16.1 Introduction
Based on the formulation discussed in the above sections, the computer coding is
developed to solve the different kinds of numerical problems incorporating linear/non-
linear analysis. In case of displacement field based on constant transverse component,
numerical code was developed in FORTRAN 90 environment. While considering the
quadratic variation of transverse component, the problem of numerical instability occurs in
FORTRAN 90, and hence further programs are developed in more versatile numerical
platform MALAB (R2013b). The developed computer program will be able to carry out
the linear/non-linear analysis of functionally graded structures with different combinations
of boundary and loading conditions.
3.16.2 Application domain

To develop the generalized computer coding, different material combinations, loading and
boundary conditions have been accomplished in the language. The developed computer
programs have been utilized to perform the following analyses.
 Linear static analysis: This module is used for the evolution of displacement and
stresses at points of interest by considering single/sandwich FGM plate/shells under
thermo-mechanical environment.
 Free/forced vibration analysis: This module is used to determine the fundamental
frequency of the FGM plate/shell having single/sandwich layers and variation of
displacement with respect to time under the application of dynamic pressure.
131
 Geometric non-linear analysis: This module is used to obtain the load-deflection
equilibrium path of FGM plate/shell under the assumption of large displacements.
 Post buckling analysis: This module is used to obtain the load-deflection plot in
the post buckling region by means of direct iteration process.
3.17 SUMMARY
The prime objective of this chapter is to present the brief mathematical background of the
linear/non-linear FE formulation. In view of giving the outline about the concept involved
in FGMs, the homogenization and different methods available for estimation of effective
mechanical properties have been presented in sections 3.2 and 3.3, respectively. The
geometrical configuration for single/sandwich layer FGM plate/shell along with the power
law distribution depending on four parameters have been described in section 3.4 and 3.5,
respectively. In section 3.6, the procedure for the skew boundary transformation has been
elaborated followed by the solution of equation for thermal analysis in section 3.7. Under
sections 3.8 and 3.9, various assumptions made in the FGM plate/shell analysis and
material modeling, respectively, adopted in the present research work have been provided.
The two different forms of displacement field assumed in the work and the governing
equation for bending analysis have been discussed under section 3.10. Next, the governing
equations for dynamic and buckling problems are derived in section 3.11 and 3.12,
respectively. Following section 3.12, the procedure implemented for geometrically non-
linear and post buckling problem has been briefed. Finally, in section 3.15, different types
of boundary restrains assumed in the study are given followed by the computer
implementation of the present FE formulation.
Based on the above FE formulation, different numerical problems are solved for
various combinations of material and geometrical parameters which are included in
subsequent chapter.
132
CHAPTER 4
RESULTS AND DISCUSSIONS
4.1 INTRODUCTION
In the present chapter, many results are presented in the form of figures and tables for
functionally graded plates and shells with/without skew constituted of single and/or multi
layers. While solving numerical examples, different combinations of material properties
and suitable values for volume fraction index are considered to represent the isotropic and
composite plate/shells for various boundary conditions and loadings. Initially, to ensure the
efficiency of the present C0 based linear/non-linear FE formulation, few results are
generated for FGM plates/shells having different boundary conditions without skew. These
results are compared with the results available in the literatures based on other shear
deformation theories. Once the efficiency of the formulation to analyze the graded
structures has been ensured, attempts have been made to generate new results for FGM
single/multilayer plates/shells with/without skew by different power law distributions.
Most of the works available in the literatures utilize the FSDT theory to predict the
responses under various operating conditions. When the transverse deflections are
estimated in the order of shell thickness, the results based on FSDT may lead to
considerable error in the analysis. Hence in the present work, FGM shells are analyzed
based on higher order theory by satisfying the shear free boundary conditions at the top
and bottom of the shell by including the realistic parabolic profile for the shear
deformation. Deflection, free vibration frequency and buckling loads are calculated under
different combinations of gradient parameters leading to conventional, symmetric and
asymmetric FGM configurations.
To eliminate the interface problems encountered in the conventional sandwich
structures, the functional grading concept was employed in sandwich layers by
incorporating core layer with either FGM or isotropic material. In the present formulation,
strain-displacement relations are considered according to Sander’s shell theory including
all three radii of curvatures (Section 3.10.1.2 of Chapter 3). Moreover, to incorporate the
influence of thickness of each layer with respect to overall thickness of the plate/shell,
different kinds of FGM sandwich plates/shells are accomplished in the present research
work.
133
To trace the load-equilibrium path in non-linear range, Green-Lagrange
relationship is incorporated in the strain formulation. In most of the studies, von Kármán
assumptions are imposed on Green-Lagrange equations to arrive for the final form of
strain-displacement relations. In the present work, all the non-linear terms present in the
Green-Lagrange equations are considered in the post buckling analysis to account for the
effect of full non-linearity. In the published literatures, the post buckling response of FGM
plates are available by assuming single layer and no results are located in the literature for
FGM sandwich plates. In order to accomplish new results, the present FE formulation has
been extended to study the post buckling behavior of FGM sandwich plates.
Various types of analyses performed in the present research work on FGM skew
plate/shells incorporating single/multi layer concept are listed below.
 Static analysis of FGM skew plates/shells under mechanical and thermal loading
conditions for different material constituents, volume fraction index, skew angles,
boundary conditions etc.
 Free vibration of FGM skew plates/shells; and also the forced vibration response of
these structures under mechanical and thermal loading conditions.
 Thermal vibration and buckling of FGM skew plates by various combinations of
ceramic and metal constituents with prescribed temperature values at the plate
boundaries.
 Static, free vibration and buckling of FGM plates based on four-parameter power
law distribution to calculate responses due to various combinations of material
profiles in the structural geometry.
 Static and free vibration of FGM sandwich skew plates/shells by considering
different kinds of sandwich plates subjected to mechanical loading.
 Geometric non-linear analysis of functionally graded sandwich plates/shells by
considering different loading magnitudes and different types of sandwich layers.
 Post buckling analysis of functionally graded sandwich plates for different
amplitude ratios and sandwich types.
The above analyses considering linear/non-linear strain field are performed for
FGM skew plate/shells with single and multilayer concept. Since effective material
distribution of FGMs primarily depends upon the appropriate choice of volume fraction
index, different combinations of material profiles are considered to perform the numerical
illustrations. Such choices should accurately predict the response of both isotropic and
composite plates in efficient ways. The effective properties of various ceramic and metal
134
components utilized in the present work to perform the numerical part are listed in Table
4.1. Various assessment studies are performed and new results are generated in the present
work considering various combinations of material properties, boundary conditions,
loading types, skew angle values and thickness schemes. To show the results
corresponding to ceramic and metal plates/shells, values of volume fraction index n=0 and
1000 are considered throughout the thesis work. Results based on higher order liner/non-
linear FE model are discussed briefly in the subsequent sections.
Table 4.1: Mechanical properties of different FGM constituents
Young’s Poisson’s Density (ρ)
Constituents modulus (E) ratio (𝜸) (kg/m3)
(GPa)
Aluminium (Al) 70.00 0.3 2707
Zirconia (Zr02) 151.00 0.3 3000
Stainless steel (SUS304) 322.27 0.24 2370
Silicon Nitride (Si3N4) 207.78 0.3177 8166
Aluminium oxide (Al203) 380.00 0.3 2707
Silicon Carbide (SiC) 427.00 0.17 3210
4.2 STATIC ANALYSIS OF FGM SKEW PLATE/SHELL SUBJECTED TO

MECHANICAL AND THERMAL LOADING
Under this sub section, FGM plates and shells incorporating skew boundary are
analyzed under mechanical (transverse) and thermal conditions. Parameters like thickness
ratio, aspect ratio, radius-thickness ratio, volume fraction index, skew angle, loading and
boundary conditions are also accomplished for comparison of results. The computed
deflections and stresses at the center are compared with the results available in the
literature based on FSDT and other HSDT models. Unless otherwise stated, the following
non-dimensional quantities are considered in the forthcoming examples.
w
Deflection under uniformly distributed loading: w 
h
100 Em h3
Deflection under sinusoidal loading: w 
12a 4 (1   m 2 )q0
q
Loading parameter: q0 
Em h 4
z
Thickness co-ordinate: z 
h
 xx
Axial stresses for plate:  xx 
q
 xx R 2
Axial stresses for shell:  xx 
qh 2
135
4.2.1 Convergence and comparison of results for Al/ZrO2 and Al/SiC
FGM plates under uniformly distributed and sinusoidal loading
conditions
To ensure the validity of the present FE formulation, four examples from the existing
literatures are chosen and the results of comparison are shown in the form of tables.
Different combinations of material and geometric conditions are considered to perform the
analysis.
Example 1: In this example, central deflections of Al/ZrO2 FGM plate are
presented for three types of thickness ratio a/h=5, 10 and 20 having different combinations
of boundary conditions (SSSS, CCCC, CFCF and SCSC). The restrained nodal unknowns
for the assumed boundary conditions are already presented in section 3.15.1 of Chapter 3.
The magnitude of volume fraction index (n) is varied from ceramic to metal part according
to Eq. (3.4.1) to show its influence on deflection parameter. Based on the convergence
study, it is manifested that 8x8 mesh division is adequate enough to perform the static
analysis using present nine noded isoparametric Lagrangian finite element. Hence for
functionally graded plate without any skew alignment, an 8x8 mesh is adopted in all the
cases. Percentage of difference observed between the present results and results reported
by Talha and Singh (2010) are also included in Table 4.2.
In Table 4.2, deflection results obtained are compared with Ferriera et al. (2005),
and Qian et al. (2004) which are based on higher order theory by using mesh free methods;
and Talha and Singh (2010) based on TSDT studied using finite element method. For
thickness ratio a/h=5, the present results are compared with available literature results and
quite close agreement among the results is obvious. In addition, other cases of thickness
ratio (a/h=10 and 20) are also shown as a second part in Table 4.2.
From all the combinations shown, it is evident that rise in volume fraction index
(n) tends to elevate the deflection of the plate irrespective of other parameters (thickness
ratio (a/h) and boundary condition) considered. The reason contributed to the above
statement is the less stiffness offered by metal plate (Al) compared to ceramic plate (ZrO2).
As expected, the plate with simply supported boundary condition endures large amount of
central deflection compared to other types of boundary conditions (CCCC, CFCF and
SCSC). Ascending thickness ratio (a/h) from 5 to 20 results in high deflection is the
another observation in Table 4.2.
136
Table 4.2: Convergence and comparison of results of square Al/ZrO2 plate having different
thickness ratio (a/h) and boundary conditions
Boundary n
a/h References
conditions Ceramic 0.5 1 2 Metal
SSSS Present (4x4) 0.02480 0.03291 0.03670 0.04014 0.05349
Present (6x6) 0.02481 0.03293 0.03673 0.04017 0.05353
Present (8x8) 0.02482 0.03294 0.03674 0.04018 0.05353
Present (12x12) 0.02482 0.03294 0.03674 0.04018 0.05353
Ferreira et al. (2005a) 0.02477 0.03293 0.03666 0.04009 0.05343
Qian et al. (2004) 0.02436 - 0.03634 0.03976 0.05253
Talha and Singh (2010) 0.0250 0.0319 0.0358 0.0393 0.0541
% of difference 0.72 3.3 2.6 2.2 1.05
CCCC 0.0109 0.0143 0.0161 0.0179 0.0235

CFCF 0.0569 0.0752 0.0840 0.0921 0.1226
SCSC 0.0158 0.0208 0.0233 0.0257 0.0340
10 SSSS 0.3459 0.4618 0.5131 0.5559 0.7461

CCCC 0.1216 0.1615 0.1800 0.1965 0.2623
CFCF 0.7625 1.0164 1.1302 1.2263 1.6449
SCSC 0.1920 0.2556 0.2845 0.3097 0.4142
20 SSSS 5.3187 7.1133 7.8947 8.5297 11.4734

CCCC 1.7129 2.2870 2.5409 2.7525 3.6950
CFCF 11.5336 15.4076 17.1075 18.4957 24.8796
SCSC 2.8063 3.7496 4.1640 4.5056 6.0537
Example 2: As second validation example, non-dimensional central deflection of

Al/ SiC plate is considered and comparison results are presented in Table 4.3. Once again,
it is observed that the present results are quite close to the published results by Ferreira et
al. (2005a). Various conclusions drawn from Table 4.2 related to volume fraction index
(n), thickness ratio (a/h) and boundary conditions hold true also for Table 4.3.
Example 3: In this example, central deflection is evaluated for Al/SiC FGM plate
x y
subjected to a sinusoidal pressure variation of magnitude q  q0 sin sin having
a b
thickness ratio (a/h) =5 and presented in Table 4.4. In this example, an effort has been
made to observe the influence of aspect ratio (b/a) on dimensionless deflection of FGM
plate at the center having CFCF boundary for several values of volume fraction index (n).
The present results based on HSDT are compared with the results of Talha and Singh
(2010) based on TSDT considering quadratic variation of transverse displacement. It is
observed that the present results are in good agreement with the above mentioned
published results.
137
Table 4.3: Convergence and comparison of results of square Al/SiC plate having different
thickness ratio (a/h) and boundary conditions
Boundary n
a/h Reference
conditions Ceramic 0.5 1 2 Metal
SSSS Present 0.00911 0.01871 0.02390 0.02932 0.05353
Ferreira et al. (2005a) 0.00909 0.01871 0.02381 0.02903 0.05343
% of difference 0.22 0 0.37 0.99 0.19
CCCC 0.00385 0.00786 0.01022 0.01303 0.02351
CFCF 0.01979 0.04140 0.05334 0.06623 0.12259
SCSC 0.00565 0.01158 0.01493 0.01875 0.03399
10 SSSS 0.12957 0.26686 0.33750 0.40501 0.74645

CCCC 0.04479 0.09198 0.11729 0.14333 0.26238
CFCF 0.27062 0.56953 0.72651 0.87988 1.64583
SCSC 0.07123 0.14646 0.18613 0.22577 4.14418
20 SSSS 2.00715 4.13755 5.21742 6.22113 11.4974

CCCC 0.64358 1.32550 1.67585 2.01002 3.70622
CFCF 4.12776 8.69888 11.0536 13.2733 24.9134
SCSC 1.05652 2.17682 2.74907 3.28893 6.07035
Table 4.4: Convergence and comparison of results of simply supported Al/SiC FGM plate
having different aspect ratios (b/a) (a/h=5.0)
n
b/a Reference
Ceramic 0.5 1 2 5 Metal
0.5 Present 0.1187 0.1559 0.1749 0.1940 0.2164 0.2308
Talha and Singh (2010) 0.1153 0.1447 0.1626 0.1814 0.2026 0.2158
% of difference 2.95 7.74 7.56 6.95 6.81 6.95
1.0 Present 0.4780 0.6324 0.7068 0.7760 0.8556 0.9156

Talha and Singh (2010) 0.5189 0.6556 0.7352 0.8135 0.8966 0.9545
% of difference 7.88 3.54 3.86 4.60 4.57 4.08
1.5 Present 0.9262 1.2286 1.3704 1.4986 1.6452 1.7618

Talha and Singh (2010) 0.9742 1.2343 1.3832 1.5254 1.6724 1.7804
% of difference 4.93 7.74 0.93 1.76 1.62 1.04
2.0 Present 1.3468 1.7888 1.9938 2.1758 2.3836 2.5546

Talha and Singh (2010) 1.3412 1.7016 1.9062 2.0984 2.2948 2.4430
% of difference 0.42 5.12 4.60 3.69 3.87 4.57
Since, the variation of transverse displacement is considered in case of results

reported by Talha and Singh (2010), in most of the cases, results reported in literature are
higher compared to present results. It is visualized that, as the value of aspect ratio (b/a)
approaches from 0.5 to 2.0, the deflection parameter enhances for both isotropic and FGM
plates.
Example 4: Non-dimensional normal stresses (σxx) at the centre (calculated at top
and bottom) of Al/ZrO2 plate for different values of volume fraction index (n) is presented
in Table 4.5. It can be seen that, at the top surface, the maximum compressive stress occurs
for FGM plate having value of n = 2.0, followed by other FGM and isotropic plates for the
value of n=1.0, 0.5 and 0. Also, the bottom surface of the plate experiences the maximum
138
tensile stress for isotropic case i.e., n= 0 followed by the FGM plates having n= 2.0, 1.0
and 0.5.
Table 4.5: Non-dimensional normal stresses at the center of square simply supported Al/ZrO2
plate for different skew angles subjected to uniformly distributed load (h=0.1 m)
n
ψ Reference 0 1.0 2.0
Top Bottom Top Bottom Top Bottom
0° Present 28.9422 -28.9422 37.5583 -22.4258 40.3474 -24.3107
Ferreira et l.(2005a) 28.650 -28.650 38.428 -17.815 45.899 -21.278
% of difference 1.02 1.02 2.26 25.88 12.09 14.25
15° Present 26.9549 -26.9549 34.9808 -20.8850 37.5798 -22.6409
30° Present 21.4786 -21.4786 27.8782 -16.6392 29.9535 -18.0399
45° Present 13.8664 -13.8664 18.0037 -10.7373 19.3534 -11.6458
60° Present 6.3131 -6.3131 8.2019 -4.8814 8.8366 -5.3052
4.2.2 Convergence and comparison of results for Al/ZrO2 cylindrical shell

subjected to mechanical load
Example 1: To validate, the present results for FGM cylindrical shell panel with a = b =
1m, Rx = 1 m, and h = 0.01 m is considered. A mechanical load of magnitude q =1.0x106
N/m2 is applied in the transverse direction at the top of the plate. Table 4.6 shows the
efficiency and convergence of the present formulation in comparison with the source paper
by Zhao et al. (2009c) based on FSDT. An excellent agreement between the results may be
observed in Table 4.6. Based on the present mesh refinement, it is decided that 12x12
mesh division is adequate for the bending analysis of FGM shells. Here, the value of
volume fraction index n is chosen from 0 to 5.
Table 4.6: Convergence and comparison of results of square Al/ZrO2 FGM shell for different
boundary conditions and volume fraction index
Boundary n
Reference
conditions 0 0.2 0.5 1 2 5
SSSS Present (4x4)1 0.04235 0.04747 0.05367 0.06027 0.06612 0.07192
Present (6x6) 0.04265 0.04781 0.05405 0.06069 0.06657 0.07240
Present (8x8) 0.04272 0.04789 0.05414 0.06079 0.06668 0.07251
Present (12x12) 0.04275 0.04792 0.05419 0.06084 0.06673 0.07256
Present (16x16) 0.04275 0.04793 0.05419 0.06085 0.06674 0.07257
Zhao et al. (2009c) 0.04267 0.04807 0.05425 0.06072 0.06658 0.07235
% of error 0.18 0.29 0.11 0.21 0.24 0.30
CCCC Present (16x16) 0.01374 0.01539 0.01739 0.01953 0.02146 0.02341

Zhao et al. (2009c) 0.01347 0.01516 0.01711 0.01915 0.02102 0.02289
% of error 2.00 1.52 1.64 1.98 2.09 2.27
CSCS Present (16x16) 0.02161 0.02423 0.02740 0.03077 0.03375 0.03668

Zhao et al. (2009c) 0.02122 0.02391 0.02700 0.03022 0.03310 0.03593
% of error 1.84 1.34 1.48 1.82 1.96 2.09
Three different boundary conditions namely, simply supported (SSSS), clamped

(CCCC) and clamped-simply supported (CSCS) are considered. A least error of about
139
0.11% is reported for simply supported boundary with n=0.5 and maximum error was
noted to be around 2.27% with n= 5.0 for clamped boundary condition. The discrepancy
between the two results may be due to different theories involved to refer the kinematics
field and the various solution strategies followed. It is to be noted that different values of
volume fraction index n have no pronounced influence on the convergence rate for FGM
shell. Also, the shell with clamped-clamped (CCCC) boundary undergoes less deflection
followed by clamped-simply supported (CSCS) and simply supported (SSSS) boundaries.
4.2.3 Non-dimensional central deflection and normal stresses for FGM

plate under thermal loading
Table 4.7: Effect of thermal loading on non-dimensional normal stresses of simply supported
Al/ZrO2 square plate for different volume fraction index and thickness ratio (a/h)
n
a/h T top (°C) Ceramic 0.5 1 2 Metal
5 20 0 -4.231 x10-4 -6.111 x10-4 -6.743 x10-4 0

50 -7.183x10-4 -1.829 x10-3 -2.228 x10-3 -2.373 x10-3 -1.652 x10-3
100 -1.915 x10-3 -4.139 x10-3 -4.923 x10-3 -5.205 x10-3 -4.405 x10-3
200 -4.309 x10-3 -8.761 x10-3 -1.031 x10-2 -1.087 x10-2 -9.912 x10-3
300 -6.704 x10-3 -1.338 x10-2 -1.570 x10-2 -1.653 x10-2 -1.542 x10-2
10 20 0 -1.769 x10-3 -2.445 x10-3 -2.697 x10-3 0
50 -2.873 x10-3 -7.312 x10-3 -8.911 x10-3 -9.493 x10-3 -6.608 x10-3
100 -7.661 x10-3 -1.665 x10-2 -1.969 x10-2 -2.082 x10-2 -1.762 x10-2
200 -1.7724 x10-2 -3.503 x10-2 -4.124 x10-2 -4.347 x10-2 -3.965 x10-2
300 -2.682 x10-2 -5.351 x10-2 -6.280 x10-2 -6.613 x10-2 -6.167 x10-2
20 20 0 -7.072 x10-3 -9.778 x10-3 -1.079 x10-2 0
50 -1.149 x10-2 -2.924 x10-2 -3.564 x10-2 -3.797 x10-2 -2.643 x10-2
100 -3.064 x10-2 -6.619 x10-2 -7.874 x10-2 -8.327 x10-2 -7.048 x10-2
200 -6.894 x10-2 -1.400 x10-1 -1.650 x10-1 -1.734 x10-1 -1.586 x10-1
300 -1.072 x10-1 -2.140 x10-1 -2.592 x10-1 -2.645 x10-1 -2.467 x10-1
100 20 0 -1.760 x10-1 -2.434 x10-1 -2.686 x10-1 0
50 -2.861 x10-1 -7.280 x10-1 -8.873 x10-1 -9.454 x10-1 -6.586 x10-1
100 -7.630 x10-1 -1.648 -1.960 -2.073 -1.755
200 -1.717 -3.488 -4.107 -4.339 -3.948
300 -2.670 -5.327 -6.253 -6.586 -6.140
Variation of normal stresses at the center of the Al/ZrO2 plate (at top and bottom) is
demonstrated in Table 4.7. The values of thickness ratio (a/h) are taken as a/h= 5, 10, 20
and 100. When subjected to thermal load, the top and bottom of the ceramic and composite
plate experience stresses that are compression in nature, while the metal segment exhibits
tensile stresses at the top and compressive stresses at the bottom.
The effect of thermal loading on central deflection of Al/ZrO2 plate for different
thickness ratio (a/h) 5, 10, 20, 100 is shown in Table 4.8. The bottom temperature of the
140
plate is kept constant as 20°C while the temperature of the plate at the top is varied from
20° C to 300°C. Negative (upward) deflection of the plate is observed under such thermal
loading, which is just opposite to the response of the plate under mechanical loading. High
thermal expansion due to increasing temperature at the top causes the plate to deflect in
negative direction.
Table 4.8: Effect of thermal loading on non-dimensional deflection of simply supported
Al/ZrO2 square plate for different volume fraction index and thickness ratio (a/h)
n
a/h T top (°C) Ceramic 1 Metal
5 20 -0.1377 -0.1377 -0.1477 -0.1542 0.1841 -0.1479
50 -0.1376 -0.3442 -0.1448 -0.3816 0.1842 -0.3670
100 -0.1376 -0.6884 -0.1400 -0.7606 0.1843 -0.7340
200 -0.1375 -1.3769 -0.1303 -1.5185 0.1845 -1.4680
300 -0.1374 -2.0653 -0.1206 -2.2765 0.1845 -2.2021
20 -0.1377 -0.1377 -0.1477 -0.1542 0.1841 -0.1468

10 50 -0.1376 -0.3443 -0.1446 -0.3817 0.1844 -0.3671
100 -0.1374 -0.6886 -0.1394 -0.7609 0.1848 -0.7343
200 -0.1370 -1.3774 -0.1290 -1.5193 0.1857 -1.4686
300 -0.1366 -2.0661 -0.1187 -2.2776 0.1865 -2.2029
20 20 -0.1377 -0.1377 -0.1475 -0.1544 0.1841 -0.1468

50 -0.1372 -0.3446 -0.1436 -0.3823 0.1851 -0.3674
100 -0.1365 -0.6895 -0.1371 -0.7622 0.1869 -0.7352
200 -0.1350 -1.3793 -0.1242 -1.5221 0.1902 -1.4707
300 -0.1335 -2.0692 -0.1113 -2.2820 0.1936 -2.2062
100 20 -0.1377 -0.1377 -0.1402 -0.1587 0.1841 -0.1468

50 -0.1277 -0.3541 -0.1174 -0.3980 0.2070 -0.3776
100 -0.1112 -0.7148 -0.0793 -0.7970 0.2450 -0.7622
200 -0.07809 -1.4363 -0.0031 -1.5949 0.3212 -1.5314
300 -0.0450 -2.1577 -0.0730 -2.3927 0.3973 -2.3006
4.2.4 Non-dimensional central deflection and in-plane normal stress for

FGM skew plate subjected to mechanical and thermal loading
Since no reference data is available on static analysis of FGM skew plates, the problem of
an isotropic skew plate is studied first, and the obtained results are compared with the
results by Sengupta (1995), Butalia et al. (1990) and Chakrabarti et al. (2004). It is
observed that the present results are in good agreement with the published results for
100wD
different values of skew angle. Non-dimensional central deflection w 
qa 4
 Eh3 
 where D   obtained for different skew angle values (= 15°, 30°, 45°, and 60°)
 12(1   2 ) 
141
of Al/ZrO2 skew plate subjected to transverse mechanical loading is presented in Table 4.9.
As the skew angle of the plate increases from 15° to 60°, significant decrease in the
deflection is observed at the center of the plate.
The influence of volume fraction index over the deflection parameter is to increase
the deflection owing to the lower contribution of ceramic material. Therefore, it is
concluded that hike in volume fraction index (n) records more deflection and this
observation is irrespective of the nature of plate boundary. In Table 4.10, variation of
normal stresses at the top and bottom of the Al/ZrO2 skew plate is demonstrated. As
anticipated, the plate with higher skew angle (= 60°) corresponds to stresses that are low
in magnitude compared to other cases of skew angles (= 15°, 30° and 45°). Further, the
top surface of the FGM skew plate experiences tensile stresses, while the bottom surface of
the FGM skew plate corresponds to compressive nature of stresses.
Table 4.9: Non-dimensional central deflection of simply supported Al/ZrO2 square plate for
different volume fraction index and skew angles subjected to mechanical loading
 n
Ceramic 0.5 1 2 5 Metal
15° 2.660 3.6082 4.0029 4.3201 4.6773 5.8157
30° 1.8799 2.5154 2.7908 3.0216 3.2628 4.0552
45° 0.9268 1.2394 1.3753 1.4856 1.6099 1.9991
60° 0.2572 0.3437 0.3815 0.4125 0.4474 0.5548
Further to incorporate the effect of thermal field on deflection and normal stresses,
different cases of temperature field are considered and the results obtained are presented in
Tables 4.11 and 4.12. In this case, the bottom temperature of the plate is kept constant at
20°C and the top temperature is varied from 20°C to 100°C.
Table 4.10: Non-dimensional normal stresses at the centre of simply supported Al/ZrO2
square plate for different volume fraction index and skew angles subjected to mechanical
loading
` n
Ceramic 1 Metal
15° 1.07x104 -1.07x104 1.39x104 -8.30x103 2.31x104 -1.07x104
30° 8.46x103 -8.46x103 1.09x104 -6.56x103 1.82x104 -8.46x103
45° 5.29x103 -5.29x103 6.85x103 -4.10x103 1.14x104 -5.29x103
60° 2.21x103 -2.21x103 2.87x103 -1.71x103 4.77x103 -2.21x103
As observed in Table 4.7, FGM plate exhibits negative deflection parameter when
subjected to the above thermal field. Further, FGM plate with skew angle 60° shows
minimum deflection while the plate having skew angle 15° records maximum deflection
parameter. In addition, it is also observed that when the temperature of the plate at top
equals the bottom temperature there will not be any deflection for isotropic cases under
different skew angles. From Table 4.12, it can be concluded that pure ceramic, pure metal
142
and FGM plate (n=1.0) records compressive stresses, except for the case of metal segment
(at top) with skew angle 15°.
Table 4.11: Non-dimensional central deflection of simply supported Al/ZrO2 square plate for
different volume fraction index and skew angles subjected to thermal field
T top n
 (°C) Ceramic 0.5 1 2 Metal
15° 20 0 -0.6564 -0.9077 -1.0013 0
50 -1.0667 -2.7144 -3.3084 -3.5244 -2.4528
100 -2.8445 -6.1445 -7.3095 -7.7297 -6.5429
30° 20 0 -0.5133 -0.7099 -0.7833 0

50 -0.8344 -2.1227 -2.5875 -2.7574 -1.9191
100 -2.2251 -4.8052 -5.7169 -6.0474 -5.1176
45° 20 0 -0.3154 -0.4362 -0.4817 0

50 -0.5129 -1.3040 -1.5900 -1.6956 -1.1797
100 -1.3678 -2.9518 -3.5129 -3.7188 -3.1459
60° 20 0 -0.1330 -0.1840 -0.2033 0

50 -0.2164 -0.5499 -0.6706 -0.7156 -0.4977
100 -0.5771 -1.2447 -1.4816 -1.5695 -1.3272
Hence, from the static study of FGM plate by considering various material and geometric
properties, the following conclusions can be drawn. First, when the volume fraction index
increases the deflection of the plate increases both under mechanical and thermal loading
condition. When subjected to thermal load, the plate shows negative deflection and
increase in the skew angle reduce the deflection of isotropic and FGM plate.
Table 4.12: Variation of non-dimensional normal stresses at the centre of simply supported
Al/ZrO2 square plate for different volume fraction index and skew angles subjected to
thermal loading
n
T top
 (°C)
Ceramic 1 Metal
20 -0.1706x10 -0.1706x108
8
-0.1867x108 -0.1864x108 0.1684x10 8
-0.1819x108
50 -0.1779x108 -0.4192x108 -0.1980x108 -0.4582x108 0.1515x10 8
-0.4469x108
15°
100 -0.1902x108 -0.8335x108 -0.2169x108 -0.9112x108 0.1235x10 8
-0.8887x108
20 -0.2250x108 -0.2250x108 -0.2521x108 -0.2347x108 0.4323x107 -0.2400x108

50 -0.2504x108 -0.5372x108 -0.2993x108 -0.5709x108 -0.1508x107 -0.5729x108
30°
100 -0.2927x108 -0.1058x109 -0.3779x108 -0.1131x109 -0.1123x108 -0.1128x109
20 -0.3050x108 -0.3050x108 -0.3419x108 -0.3096x108 -0.1406x108 -0.3252x108

50 -0.3483x108 -0.7191x108 -0.4250x108 -0.7501x108 -0.2402x108 -0.7668x108
45°
100 -0.4205x108 -0.1409x109 -0.5634x108 -0.1484x109 -0.4062x108 -0.1503x109
20 -0.3946x108 -0.3946x108 -0.4371x108 -0.3968x108 -0.3467x108 -0.4207x108

60° 50 -0.4505x108 -0.9305x108 -0.5460x108 -0.9630x108 -0.4754x108 -0.9921x108
100 -0.5438x108 -0.1824x109 -0.7275x108 -0.1907x109 -0.6899x108 -0.1944x109
143
4.2.5 Non-dimensional central deflection and normal stresses for FGM
skew shell under mechanical and thermal loading
To analyze FGM shells under static loading, different shell forms subjected to mechanical
and thermal loading are considered. To execute thermal analysis, temperature dependent
material properties are incorporated. Unless otherwise specified, the geometric details of
the shell: a = b = 1m, Rx = 1m and h = 0.01m are considered. A mechanical load of
magnitude q =1.0x106 N/m2 is applied in the transverse direction.
4.2.5.1 Non-dimensional central deflection and in-plane normal stress for

FGM skew cylindrical shell under mechanical loading
In Table 4.13, influence of skew angle on central deflection of Al/ZrO2 cylindrical shell
(Rx=R, Ry=Rxy=∞) is demonstrated for various types of support conditions. The assumed
value of skew angle ranges from 15° to 60° and several values of volume fraction index
(n=0 to10) are also chosen for the present purpose. It is noticed that hike in skew angle
tends to decline the deflection irrespective of the value of volume fraction (n). Another
important observation from Table 4.13 is that larger volume fraction (n) of the cylindrical
shell increase the deflection parameter. Low stiffening property contributed by the metal
segment may be the cause for this type of behaviour. Shell with SSSS boundary condition
shows large amount of deflection than that of shell with CCCC and SCSC boundary
conditions. This observation is not surprising, because of the high rigidity offered by
clamped shell compared to other type of boundary conditions.
Table 4.13: Non-dimensional central deflection of square Al/ZrO2 cylindrical shell for
different skew angles and boundary conditions (h=0.01m)
Boundary n
conditions

0 0.2 0.5 1 2 5 10
SSSS 15° 0.04264 0.04792 0.05430 0.06096 0.06656 0.07184 0.07681
30° 0.02993 0.03363 0.03810 0.04278 0.04672 0.05044 0.05393
45° 0.01508 0.01694 0.01919 0.02155 0.02355 0.02546 0.02722
60° 0.00440 0.00493 0.00559 0.00627 0.00687 0.00745 0.00797
CCCC 15° 0.01364 0.01531 0.01734 0.01947 0.02131 0.02310 0.02470

30° 0.00943 0.01058 0.01198 0.01346 0.01474 0.01599 0.01710
45° 0.00473 0.00531 0.00600 0.00674 0.00740 0.00805 0.00861
60° 0.00145 0.00162 0.00183 0.00206 0.00227 0.00248 0.00266
SCSC 15° 0.02237 0.02513 0.02846 0.03196 0.03495 0.03781 0.04043

30° 0.01558 0.01749 0.01981 0.02224 0.02434 0.02636 0.02819
45° 0.00792 0.00888 0.01005 0.01129 0.01237 0.01343 0.01436
60° 0.00245 0.00274 0.00310 0.00349 0.00383 0.00418 0.00447
144
Non-dimensional in-plane normal stress values at the top and bottom surface of
Al/Al2O3 cylindrical panel subjected to transverse mechanical load is exhibited in Table
4.14. The bottom of isotropic and FGM shell experiences the stresses that are compressive
in nature, while the top of the shell experiences the stresses that are tensile in character.
Further, increase in skew angle tends the shell to experience minimum stresses (either
compressive or tensile in nature) both at the top and bottom surfaces. Also, due to increase
in volume fraction index (n), the shell is subjected to maximum stress (either compressive
or tensile in nature) at bottom and top.
Table 4.14: Non-dimensional normal stresses of simply supported Al/ZrO2 square cylindrical
shell for different skew angles and volume fraction index (h=0.01m)
n
 Ceramic 1 Metal
15° 0.1104e09 -0.79508e08 0.16848e09 -0.40741e08 0.50517e08 -0.79508e08
30° 0.90266e08 -0.61881e08 0.13845e09 -0.31937e08 0.41290e09 -0.61881e08
45° 0.60161e08 -0.40295e08 0.92840e08 -0.20950e08 0.27519e09 -0.40295e08
60° 0.26999e08 -0.18978e08 0.41680e08 -0.98531e07 0.12350e09 -0.18978e08
4.2.5.2 Non-dimensional central deflection of Al/Al2O3 spherical, hypar

and hyperbolic paraboloid skew shell subjected to mechanical loading
In Figure 4.1, the superiority of Al/Al2O3 FGM shell having different plan views is
demonstrated under bending. Bending performance of three kinds of shell types namely
spherical (Rx=Ry=R, Rxy=∞), hyperbolic paraboloid (Rx=-Ry) and hypar (Rx=Ry=∞) are
considered to generate the deflection results. It is to be noted that the incorporation of twist
curvature (Rxy) term in the strain part will enable the formulation to handle the special form
like hypar skew shells. Deflection results are presented for the linear variation of volume
fraction index i.e, n=1.0. Four kinds of support conditions viz., simply supported-simply
supported (SSSS), clamped-clamped (CCCC), simply supported-clamped (SCSC) and
clamped-free (CFCF) are considered in the present case. Among the different forms of
shell considered, hypar FGM shells shows lower deflection values compared to spherical
and hyperbolic paraboloid FGM skew shell.

145
0.07
0.07
Non-dimensional central deflection (w/h)

0.06
0.06 Boundary condition
SSSS
0.05 CCCC
Boundary condition 0.05
SCSC
SSSS
0.04 CFCF
CCCC
0.04
SCSC
0.03 CFCF
0.03
0.02
0.02
0.01 0.01
0.00 0.00
10 20 30 40 50 60 10 20 30 40 50 60
Skew angle (in degree) Skew angle (in degree)

(a) Spherical shell (b) Hyperbolic paraboloid
0.05
Boundary condition
SSSS
CCCC
0.04 SCSC
CFCF
0.03
0.02
0.01
0.00
10 20 30 40 50 60
Skew angle (in degree)

(c) Hypar
Figure 4.1 Influence of boundary conditions on non-dimensional central deflection of
simply supported Al/Al2O3 square shell for different skew angles subjected to mechanical
loading (h=0.01m)
4.2.5.3 Influence of R/h ratio on non-dimensional deflection of Al/Al2O3

cylindrical skew shell under mechanical loading
Figure 4.2 exhibits the influence of R/h ratio on deflection parameter of Al/Al2O3 shell
having simply supported support condition. Three types of radius-thickness ratios R/h= 50,
100 and 200 and several values of volume fraction index n=0.5, 1.0, 2.0 and 10 are chosen
to demonstrate the bending response of Al/Al2O3 skew cylindrical shell. Figure 4.2 shows
that as the radius of shell increases, the deflection value decreases for different skew
angles. Increase in volume fraction index corresponds to less stiffness and hence ensures
maximum deflection value for FGM skew plates. For the value of R/h=50 which means
thickness h=0.02m shows almost least influence on the bending response of cylindrical
shell.
146
0.16
0.20

0.12 0.16
0.12
0.08
Radius-thickness ratio
R/h=50 0.08
R/h=100 R/h=50
R/h=200 R/h=100
0.04
R/h=200
0.04
0.00 0.00
10 20 30 40 50 60 10 20 30 40 50 60
0.25 (b) Volume frcation index (n)=1
(a) Volume frcation index (n)=0.5

0.30
0.20
0.25
0.20
0.15
0.15 Radius-thickness ratio

0.10
R/h=50 R/h=50
R/h=100 0.10
R/h=100
R/h=200 R/h=200
0.05
0.05
0.00 0.00
10 20 30 40 50 60 10 20 30 40 50 60

(d) Volume frcation index (n)=2 (d) Volume frcation index (n)=10
Figure 4.2 Influence of radius-thickness ratio on non-dimensional central deflection of

simply supported Al/Al2O3 square cylindrical shell for different skew angles subjected to
mechanical loading (h=0.01m)
0.05
0.04
n=0
n=0.5
n=2
0.03 n=5
n=1000
0.02
0.01
0.00
100 200 300 400 500 600 700 800
Load parameter (q0)
Figure 4.3 Influence of load parameter on non-dimensional central deflection of

simply supported square Al/Al2O3 cylindrical shell for different volume fraction
index subjected to mechanical loading (h=0.01m)
147
Further, for R/h=200 which means h=0.005m reflects drastic bending response compared
to other cases. The consequence of load parameter (q0) on non-dimensional deflection for
Al/Al2O3 cylindrical shell is shown in Figure 4.3 for several values of volume fraction
index (n=0 to 1000). Higher value of load parameter indicates rise in deflection parameter
for all the cases of volume fraction index considered. Further, increase in the volume
fraction index (n) tends to increase the deflection, which is same as observed in preceding
examples.
4.2.5.4 Effect of thermal loading on non-dimensional deflection of

Al/Al2O3 skew cylindrical shell
Table 4.15: Non-dimensional central deflection of SUS3O4/Si3N4 square cylindrical shell for
different volume fraction index and skew angles subjected to thermal field (h=0.01m)
Ttop n
 (°C) Ceramic 0.5 1 2 5 Metal
15° 100 0.02030 0.01920 0.01940 0.02140 0.02730 0.04410
150 0.03290 0.03120 0.03160 0.03480 0.04440 0.07170
200 0.04560 0.04320 0.04370 0.04820 0.06140 0.09930
250 0.05830 0.05510 0.05590 0.06150 0.07850 0.12700
300 0.07100 0.06710 0.06800 0.07490 0.09550 0.15400
350 0.08360 0.07910 0.08020 0.08830 0.11300 0.18200
400 0.09630 0.09110 0.09230 0.10200 0.13000 0.21000
100 0.01590 0.01510 0.01530 0.01680 0.02150 0.03470

150 0.02590 0.02450 0.02480 0.02740 0.03490 0.05640
200 0.03590 0.03390 0.03440 0.03790 0.04830 0.07810
30° 250 0.04580 0.04340 0.04390 0.04840 0.06170 0.09980
300 0.05580 0.05280 0.05350 0.05890 0.07510 0.12200
350 0.06580 0.06220 0.06310 0.06940 0.08860 0.14300
400 0.07570 0.07170 0.07260 0.08000 0.10200 0.16500
100 0.00998 0.00944 0.00957 0.01050 0.01340 0.02170

150 0.01620 0.01530 0.01550 0.01710 0.02180 0.03530
200 0.02250 0.02120 0.02150 0.02370 0.03020 0.04890
45° 250 0.02870 0.02710 0.02750 0.03030 0.03870 0.06250
300 0.03490 0.03300 0.03350 0.03690 0.04710 0.07610
350 0.04120 0.03890 0.03950 0.04350 0.05550 0.08970
400 0.04740 0.04490 0.04540 0.05010 0.06390 0.10300
100 0.00432 0.00408 0.00414 0.00456 0.00582 0.00942

150 0.00701 0.00663 0.00672 0.00741 0.00946 0.01530
200 0.00971 0.00919 0.00931 0.01030 0.01310 0.02120
60° 250 0.01240 0.01170 0.01190 0.01310 0.01670 0.02710
300 0.01510 0.01430 0.01450 0.01600 0.02040 0.03300
350 0.01780 0.01680 0.01710 0.01880 0.02400 0.03890
400 0.02050 0.01940 0.01970 0.02170 0.02770 0.04480
The effect of temperature field on the bending response of SUS3O4/Si3N4 cylindrical shell
is shown in Table 4.15. The temperature dependent material properties of the constituents
are: for SUS3O4, E=207.78 Gpa, =0.318, 𝛼 = 15.3x10-6/°C, K=9.54 W/m-K; for Si3N4
E=322.27 Gpa, =0.24, 𝛼 = 7.47x10-6/°C, K=10.12 W/m-K at T=300K. The bottom
148
temperature which refers to temperature of the metal plate is kept as unchanged (i.e.,
Tm=20°C), while the top temperature which corresponds to the temperature of the ceramic
plate assumes the value from 100°C to 400°C. It is manifested that the amplification of
temperature field is to increase the deflection regardless of the skew angle of the shell.
This fact is observed due to low value of Young’s modulus of the material under high
temperatures.
4.3 FREE VIBRATION OF FGM SKEW PLATE/SHELL

Free vibration frequency study is performed in this section for different combination of
material and geometric parameters by taking different skew angle and volume fraction
index values. Also, in many cases, frequency mode shapes are plotted for different skew
angles.
4.3.1 Convergence and comparison of results of free vibration frequency
for FGM skew plate
The following non-dimensional quantities are used to represent the numerical results in
this section.
1
 a 2  c  2
For FGM plate:    
h  Ec 
For FGM shell:   a 2 12m (1   2 ) Em h2
In the above expressions, the subscripts ‘c’ and ‘m’ represent the ceramic and metal
constituents, respectively.
4.3.1.1 Comparison of frequency values of SuS3O4/Si3N4 plate
In this example, SuS3O4/Si3N4 skew plate with simply supported boundary is considered
for free vibration analysis and the results obtained are shown in Table 4.16. The
convergence of the frequency values are observed for mesh division 16x16. The results are
compared with FSDT results of Zhao et al. (2009a) and TSDT results presented by Talha
and Singh (2010). Percentage of difference between the present results and results
reported by Talha and Singh (2010) is also included in Table 4.16.
A close observation in Table 4.16 shows that, larger volume fraction index leads to
decrease in frequency parameter for all the skew angles. Another observation in Table 4.16
is that, as the skew angle of the plate elevates the frequency parameter increases and this is
valid for all the values of volume fraction index. Once the efficiency of the present FE
149
formulation to obtain free vibration frequency results has been ensured, new results are
generated for skew plate/shell considering different material and geometric parameters.
Table 4.16: Non-dimensional frequency parameter of simply supported
SuS3O4/Si3N4 square skew plate (h=0.1m)
n
ψ References
0 0.5 1.0 2.0
0° Present (16x16) 5.6821 3.9238 3.4477 3.1044
Zhao et al. (2009a) 5.6148 3.8947 3.4242 3.0813
Talha and Singh (2010) 5.6523 3.9201 3.4415 3.1062
% of difference 0.53 0.09 0.18 0.06
15° Present 7.4189 5.1061 4.4757 4.0179

30° Present 11.9181 8.1809 7.1475 6.3852
45° Present 19.4377 13.3136 11.5928 10.3033
60° Present 33.9717 23.2189 20.1583 17.8404
4.3.1.2 Non-dimensional frequency values of SUS3O4/Si3N4 skew plate
60
Skew angle 70
0°
Non-dimensional frequency parameter ( 
15°
50
30° 60
45°
60°
40 50
30 40
20 30
Skew angle
0°
20 15°
10
30°
45°
10 60°
0
1 2 3 4 5 6 1 2 3 4 5 6
Frequency mode Frequency mode
(a) Simply supported boundary condition (b) Clamped boundary condition
Skew angle
60 0°
15°
30°
50 45°
60°
40
30
20
10
1 2 3 4 5 6
Frequency mode
(c) Simply supported-clamped boundary condition
Figure 4.4 Influence of boundary conditions on frequency parameter of SuS3O4/Si3N4 square

plate for different skew angles (h=0.1m and n=1.0)
150
Next, the influence of skew angle (ψ= 0⁰, 15⁰, 30⁰, 45⁰ and 60⁰) on first six frequency
modes of SUS304/Si3N4 skew plate for different boundary conditions (SSSS, CCCC,
SCSC), aspect ratio (a/b=1.0, 2.0) and thickness ratio (h/a=0.1, 0.2) are shown in Figures
4.4, 4.5 and 4.6. From the frequency plots, it can be inferred that, imposing constrains on
the boundaries of the plate increase the frequency parameter for all the skew angles. As
expected, frequency value increases with increase in skew angle of the FGM plate. From
Figure 4.5, it is observed that frequency parameters shows fall-off tendency as the aspect
ratio rises from 1 to 2; and significant decrease is observed in frequency as the thickness
ratio changes from 0.1 to 0.2. Influence of volume fraction index (n) on frequency
parameter for several values of skew angle is exhibited in Figure 4.7.
Skew angle Skew angle

36 0°
0° Non-dimensional frequency parameter ( 
28 15°
15°
30° 32 30°
45° 45°
24
60° 60°
28
20
24
16
20
12 16
8 12
8
4
1 2 3 4 5 6 1 2 3 4 5 6
32
Skew angle
0°
28 15°
30°
45°
24
60°
20
16
12
1 2 3 4 5 6
Frequency mode
Figure 4.5 Influence of boundary condition on frequency parameter of square SUS3O4/Si3N4

plate for different skew angles (h=0.2m and n=1.0)
Two common observations are discerned from figures (Figures 4.4, 4.5, 4.6 and
4.7). The first observation is that irrespective of the boundary constraints as the skew angle
151
40
0° 0°

50
15° 15°
32 30° 30°
45° 45°
40
60° 60°
24
30
16
20
8
10
1 2 3 4 5 6 1 2 3 4 5 6
50
Skew angle
0°
15°
40 30°
45°
60°
30
20
10
1 2 3 4 5 6
Frequency mode
Figure 4.6 Influence of boundary conditions on frequency parameter of SUS3O4/Si3N4 plate

for different skew angles (b/a=2.0, h=0.1m and n=1.0)
40
Skew angle 36
0° Skew angle
36
15° 0°
30° 30 15°
32
45° 30°
60° 45°
28
60°
24
24
20 18
16
12
12
8
6
4
1 2 3 4 5 6 1 2 3 4 5 6
(a) Volume fraction index (n)=0 (b) Volume fraction index (n)= 1.0
152
36
Skew angle

0°
30 15°
30°
45°
24 60°
18
12
1 2 3 4 5 6
Frequency mode
(c) Volume fraction index (n)= 2.0
Figure 4.7 Influence of volume fraction index on frequency parameter of SUS3O4/Si3N4 plate
for different skew angles (b/a=2.0, h=0.1m and n=0)
increases, free vibration frequency results increases. Second observation is that if the value
of volume fraction index increases the stiffness of the plate decreases due to lower
Young’s modulus of metal constituent. In addition, plate with clamped boundary shows
more free vibration frequency results compared to other type of boundary conditions.
4.3.2 Thermal vibration Frequency for FGM skew plate having different
temperature dependent properties
Under this section, the natural frequency results for thermal vibration are presented for
FGM skew plates by considering different thermal properties of the FGM plate. First six
frequency modes are shown for different cases of thermal vibration. While solving
numerical examples, various parameters such as boundary condition, thickness ratio,
aspect ratio and volume fraction index are varied to sketch their influence on frequency
parameter. All the results are presented in following non-dimensional form.
 
0.5
  0 1   
2
 a2
For FGM plate:     h   
   E0

where ρ0 and E0 are the reference values of metal taken at T0=300K.
4.3.2.1 Validation of non-dimensional free vibration frequency values of

SUS3O4/Si3N4 plate
Since, no results are available on thermal vibration of plates having skew boundary, the
results for rectangular FGM plates published by Huang and Shen (2004) are used to
validate the results based on the present FE model. In this example, Young’s modulus and
153
thermal expansion are considered as temperature dependent; thermal conductivity and
density are considered as temperature in-dependent, the temperature dependent properties
of different FGM constituents are reported in Table 4.17.
Table 4.17: Temperature dependent properties for different FGM constituents at T=300K
(Reddy and Chin 1998)
Constituents Properties P0 P-1 P1 P2 P3 P(T=300K)
Zr02 E (Pa) 244.27e+09 0 -1.371e-03 1.214e-06 -3.681e-10 168.063e+09
α (1/K) 12.766e-06 0 -1.491e-03 1.006e-05 -6.778e-11 18.591e-06
Ti-6Al-4V E (Pa) 122.56e+09 0 -4.586e-04 0 0 105.698e+09

α (1/K) 7.5788e-06 0 6.638e-04 -3.147e-06 0 6.941e-06
Si3N4 E (Pa) 348.43e+09 0 -3.070e-04 2.160e-07 -8.946e-11 322.2715e+09

α (1/K) 5.8723e-06 0 9.095e-04 0 0 7.4746e-06
SUS304 E (Pa) 201.04e+09 0 3.079e-04 -6.534e-07 0 207.7877e+09

α (1/K) 12.330e-06 0 8.086e-04 0 0 15.321e-06
A square Si3N4/ SUS3O4 plate having plate dimension a = b = 0.2m and thickness h
= 0.025 m is considered under thermal environment for comparison purpose. The thermal
conductivity and mass density of the constituents are: for SUS3O4, ρm= 8166 kg/m3,
km=12.04 W/m K; and for Si3N4 ρc= 2370 kg/m3, kc=9.19 W/m K. The influence of
temperature on Poisson’s ratio has no significance and hence it is assumed as constant
(γ=0.28).
The non-dimensional linear frequency (ω) for simply supported Si3N4/ SUS3O4
plate under thermal environment is reported in Table 4.18. The results obtained by Huang
and Shen (2004) are used to compare the present free vibration frequencies based on
HSDT. A perturbation technique is adopted to solve the equation of motion by Huang and
Shen (2004) and close matching between the two results is observed in Table 4.18. Also,
the present numerical example is solved for different mesh divisions and it is found that
5x5 mesh division is sufficient enough for the convergence of free vibration frequency
results with desired accuracy. Hence a mesh division of 5x5 is used to generate the
frequency results. Two types of thermal environments, where temperature of bottom
constituent (metal) equals 300 K and temperature of top constituent (ceramic) is assumed
as 400K and 600K for temperature in-dependent and temperature dependent condition are
accomplished. The estimated percentage of error between the present results and results
reported by Huang and Shen (2004) is also accomplished in Table 4.18. Maximum
percentage of difference is reported under Tc= Tm= 300K for ceramic segment, while the
minimum difference was reported for the case of Tc=400K and Tm=300K where
temperature independent properties are taken with n=0 (ceramic). However, no error was
154
observed for FGM plate with temperature dependent properties for which the bottom and
top temperature of the plate are taken as 300K and 400K, respectively.
Table 4.18: Validation table showing frequency values of Si3N4/ SUS3O4 plate for different
values of volume fraction index
T References n
Ceramic 0.5 1.0 2.0 Metal
Tc=300K Present (5x5) 13.034 8.878 7.726 6.892 5.334
Tm=300K Huang and Shen (2004) 12.495 8.675 7.555 6.777 5.405
% of error 4.31 2.34 2.26 1.70 1.31
Tc=400K Present 12.811 8.759 7.638 7.827 5.290

Tm=300K (T.I) Huang and Shen (2004) 12.397 8.615 7.474 6.698 5.311
% of error 3.34 1.67 2.19 4.05 0.04
Tc=600K Present 12.353 8.516 7.459 6.695 5.203

Tm=300K (T.I) Huang and Shen (2004) 12.213 8.425 7.305 6.523 5.104
% of error 1.15 1.08 2.11 2.64 1.94
Tc=400K Present 12.405 8.532 7.465 6.693 5.307

Tm=300K (T.D) Huang and Shen (2004) 12.397 8.615 7.474 6.693 5.311
% of error 0.06 0.96 0.12 0.00 0.08
Tc=600K Present 12.142 8.252 7.186 6.423 5.068

Tm=300K(T.D) Huang and Shen (2004) 11.984 8.269 7.171 6.398 4.971
% of error 1.32 0.21 0.21 0.39 1.95
T.I-Temperature- independent; T.D-Temperature-dependent
4.3.2.2 Non-dimensional free vibration frequency values of SuS3O4/Si3N4

skew plate
To study the influence of skew angle on free vibration frequency parameter, a square
Si3N4/SUS3O4 skew plate with thickness ratio (a/h) =10 is considered. Different boundary
conditions viz., simply supported (SSSS), clamped (CCCC) and simply supported-clamped
(SCSC) are assumed and the results are presented in Tables 4.19, 4.20 and 4.21,
respectively. Temperature of Tc= 400K and Tm=300K are applied on the top and bottom of
the plate, respectively, to present new results for FGM skew plates. From tables, it can be
inferred that, enhancing skew angle of the plate increases the frequency for all the six
modes and this type of behavior is irrespective of the value of volume fraction index (n)
considered. This is due to the fact that increase in skew angle reduces the distance between
non-skew edges and thus reducing the area and thereby increases the frequency parameter
value. Further, a close range of frequency parameter is observed for skew angles 0° and
15°, beyond which noticeable difference is discerned. This deviation is independent of the
value of volume fraction index and type of boundary condition for the analysis. It is noted
that, increasing volume fraction
155
Table 4.19: Non-dimensional frequency values for simply supported Si3N4/ SUS3O4 square
plate for different values of skew angles and volume fraction index (a/h=10)
Mode
 n
1 2 3 4 5 6
Ceramic 2.0061 4.8528 4.8528 7.4805 9.2543 9.2543
0.5 1.3756 3.3344 3.3644 5.1483 6.3693 6.3723
0° 1.0 1.2022 2.9125 2.9125 4.4991 5.5643 5.5686
2.0 1.0776 2.6100 2.6100 4.0304 4.9825 4.9865
Metal 0.8530 2.0840 2.0840 3.2201 3.9890 3.9870
Ceramic 2.1240 4.7543 5.4808 7.5202 9.7136 10.096
0.5 1.4581 3.2678 3.7680 5.1773 6.6868 6.9537
15° 1.0 1.2749 2.8547 3.2916 4.5249 5.8417 6.0765
2.0 1.1431 2.5585 2.9496 4.0537 5.2303 5.4405
Metal 0.9053 2.0428 2.3565 3.2386 4.1868 4.3519
Ceramic 2.5542 5.1870 6.9332 8.0161 9.9596 11.208
0.5 1.7585 3.5693 4.7719 5.5237 6.9207 7.7743
30° 1.0 1.5392 3.1192 4.1695 4.8291 6.0239 6.7724
2.0 1.3806 2.7958 3.7355 4.3260 5.3316 6.0061
Metal 1.0948 2.2338 2.9878 4.4569 4.3087 4.8462
Ceramic 3.6349 6.4917 9.6328 9.7160 10.1312 13.260
0.5 2.5108 4.4738 6.6475 6.7682 6.9805 9.1631
45° 1.0 2.1998 3.9107 5.8130 5.8982 6.0986 8.0147
2.0 1.9736 3.5047 5.2054 5.2267 5.4597 7.1702
Metal 1.5677 2.8051 4.1637 4.2141 4.3754 5.7330
Table 4.20: Non-dimensional frequency values for clamped Si3N4/ SUS3O4 square plate for
different values of skew angles and volume fraction index (a/h=10)
Mode
 n
1 2 3 4 5 6
Ceramic 3.4690 6.6929 6.6929 9.4163 11.3369 11.445
0.5 2.3888 4.6153 4.6153 6.4762 7.8247 7.9011
0° 1.0 2.0860 4.0300 4.0300 5.6718 6.8302 6.8973
2.0 1.8668 3.6025 2.6025 5.0671 6.0975 6.1575
Metal 1.4906 2.8829 2.8829 4.0596 4.8900 4.9372
Ceramic 3.6667 6.6507 7.4396 9.5553 11.8853 12.353
0.5 2.5255 4.5868 5.1316 6.5934 8.2046 8.5294
15° 1.0 2.2054 4.0050 4.4806 5.7564 7.1613 7.4451
2.0 1.9734 3.5798 4.0043 5.1419 6.3917 6.6448
Metal 1.5761 2.8648 3.2057 4.1199 5.1272 5.3296
Ceramic 4.3621 7.3539 9.1829 10.3505 13.6765 13.772
0.5 3.0068 5.0746 6.3381 7.1463 9.4779 9.5115
30° 1.0 2.6257 4.4305 5.5331 6.2382 8.2556 8.3003
2.0 2.3484 3.9581 4.9412 5.5689 7.3231 7.4028
Metal 1.8769 3.1691 3.9595 4.4641 5.9134 5.9434
Ceramic 5.9966 9.2809 12.591 12.8447 15.4032 16.404
0.5 4.1392 6.4117 8.7041 8.8735 10.7022 11.350
45° 1.0 3.6140 5.5963 7.5955 7.7429 9.3196 9.9030
2.0 3.2289 4.9933 6.7713 6.9045 8.2639 8.8197
Metal 2.5839 4.0027 5.4338 5.5428 6.6769 7.0822
index from ceramic to metal segment reduces the frequency parameter considerably. The
reason owing to the observation is that the increase in volume fraction index corresponds
to the less ceramic portion (i.e., bottom of the plate) and hence represents the less stiffness
of the plate. Further, among the different boundary conditions assigned, highest frequency
value is observed for CCCC plate and lowest frequency is reported for SSSS FGM plate
Imposing constrains on the boundary of the plate increases the frequency of the plate.
156
Table 4.21: Non-dimensional frequency values for simply supported-clamped Si3N4/ SUS3O4
square plate for different values of skew angles and volume fraction index (a/h=10)
Mode
 n
1 2 3 4 5 6
Ceramic 2.6811 5.7285 5.7647 8.4221 10.2993 10.325
0.5 1.8427 3.9444 3.9694 5.8041 7.0982 7.1179
1.0 1.6095 3.4452 3.4673 5.0700 6.1975 6.2157
0°
2.0 1.4415 3.0838 3.1036 4.5356 5.5403 5.5569
Metal 1.1475 2.4643 2.4801 3.6285 4.4402 4.4515
Ceramic 2.8234 5.6934 6.3988 8.5206 10.8052 11.168

0.5 1.9418 3.9212 4.4074 5.8728 7.4485 7.6998
1.0 1.6965 3.4254 3.8497 5.1299 6.5035 6.7224
15°
2.0 1.5195 3.0661 3.4452 4.5888 5.8132 6.0075
Metal 1.2098 2.4501 2.7547 3.6719 4.6596 4.8167
Ceramic 3.3437 6.2614 7.9522 9.1659 11.5464 12.557

0.5 2.3035 4.3159 5.4813 6.3212 8.0209 8.6608
1.0 2.0135 3.7706 4.7877 5.5214 6.9841 7.5608
30°
2.0 1.8035 3.3744 4.2829 4.9373 6.1887 6.7543
Metal 1.4370 2.6976 3.4282 3.9531 4.9940 5.4192
Ceramic 4.5859 7.8831 11.092 11.283 11.7433 14.852

0.5 3.1667 5.4409 7.6525 7.7839 8.1784 10.264
45° 1.0 2.7697 4.7536 6.6793 6.7973 7.1313 8.9705
2.0 2.4805 4.2511 5.9650 6.0753 6.3227 8.0066
Metal 1.9785 4.4031 4.7907 4.8722 5.0881 6.4167
4.3.2.3 Influence of thermal field on non-dimensional frequency values of

SUS3O4/Si3N4 skew plate
Table 4.22: Influence of thermal field on non-dimensional frequency values of Si3N4/
SUS3O4 square plate for different values of skew angles and boundary conditions subjected to
thermal field (h=0.1m)
Boundary ∆T (C)
 conditions 0 400 800 1200
CCCC 2.1079 2.0343 1.9574 1.8771
SSSS 1.2357 1.1319 1.0176 0.8888
0°
SCSC 1.6367 1.5476 1.4527 1.3506
CCCC 2.2274 2.1538 2.0772 1.9972

SSSS 1.5622 1.4924 1.4164 1.3333
15°
SCSC 2.0345 1.9690 1.8935 1.8164
CCCC 2.6477 2.5742 2.4982 2.4194

SSSS 1.5622 1.4924 1.4164 1.3333
30°
SCSC 2.0344 1.9660 1.8935 1.8164
CCCC 3.6363 3.5624 3.4867 3.4092

45° SSSS 2.2105 2.1773 2.1343 2.0817
SCSC 2.7812 2.7419 2.6945 2.6396
To infer the influence of temperature field across the thickness on frequency parameter,
Si3N4/SUS3O4 skew plate having thickness (h) 0.1 m is considered in the present section
and results are furnished in Table 4.22. Various combinations of boundary conditions
157
(CCCC, SSSS, and SCSC) are adopted and the value of volume fraction index is taken as
1.0. The temperature range of 0° - 1200° K is selected to obtain the frequency values of
skew plate for different skew angles. Two observations are visualized from the example.
First, the rise in temperature reduces the frequency of the plate for all the types of
boundary condition. This type of observation is expected, because at high temperature the
Young’s modulus of the plate becomes less, thus reducing the frequency of the plate. After
that, there is significant increase in frequency parameter as the skew angle of plate rises
from 0° to 45°.
Table 4.23: Influence of thermal field on non-dimensional frequency values of simply

supported ZrO2/ Ti-6Al-4V square plate for different values of skew angles (h=0.1m)
Mode
 ∆T (C)
1 2 3 4 5 6
100 6.6238 16.0508 16.0508 24.7913 30.6590 30.6803
300 6.2144 15.6369 15.6369 24.3646 30.2231 30.2451
0° 600 5.5438 14.9944 14.9944 23.7107 29.5569 29.5802
900 4.7800 14.3232 14.3232 23.0370 28.8755 28.9009
100 7.0262 15.7338 18.1418 24.9345 32.1866 33.4773

300 6.6531 15.3595 17.7539 24.5455 31.7831 33.0639
15° 600 6.0493 14.7787 17.1555 23.9456 31.1652 32.4348
900 5.3764 14.1722 16.5353 23.3377 30.5315 31.7953
100 8.4876 17.1955 22.9816 26.6122 33.3158 31.0082

300 8.2126 16.9253 22.6758 26.3281 33.3205 36.9234
30° 600 7.7684 16.4981 22.2046 25.8907 33.3253 36.6705
900 7.2796 16.0424 21.7176 25.4370 33.3264 36.2176
100 12.133 21.5628 32.0284 32.6081 33.6084 44.1183

300 11.997 21.4734 31.9400 32.0975 33.4177 44.0362
45°
600 11.745 21.2859 31.7710 33.0998 33.3140 43.8861
900 11.455 21.0377 31.5596 32.7674 33.6872 43.7040
In some practical cases, it is observed that buckling occurs before the plate starts vibrating
in its natural frequency. Therefore to study such situation, uniform temperature is applied
over the plate to extract its critical buckling temperature. Then the temperature below this
critical buckling temperature is applied to calculate the thermal frequency value of the
plate. Such frequency values for simply supported ZrO2/ Ti-6Al-4V skew plate is depicted
in Table 4.23. The first six frequency modes were reported under different cases of thermal
conditions and linear variation of volume fraction index is considered in the example. Free
vibration frequency results with respect to skew angle, boundary condition, and
temperature concluded from Table 4.22 hold good for Table 4.23 also.
158
4.3.2.4 Influence of aspect ratio on non-dimensional free vibration
frequency values of Si3N4/SUS3O4 skew plate
16
Skew angle

Skew angle
0°
0° 15°
14
8 15° 30°
30° 45°
45°
12
6 10
4
6
Ceramic 0.5 1 2 Metal Ceramic 0.5 1 2 Metal

Volume fraction index (n) Volume fraction index (n)
(a) Aspect ratio (a/b)=0.5 (b) Aspect ratio (a/b)=1.0
40 Skew angle
Skew angle
24 0°
0°
15° 15°
30°
35 30°
45° 45°
20
30
16 25
20
12
15

(c) Aspect ratio (a/b)=1.5 (d) Aspect ratio (a/b)=2.0
Figure 4.8 Influence of aspect ratio (a/b) on frequency parameter of simply supported
Si3N4/SUS3O4 plate for different skew angles (h=0.1m)
Variations of frequency parameters of Si3N4/SUS3O4 skew plate for several aspect ratio
(a/b= 0.5, 1.0, 1.5, and 2.0) and simply supported and clamped boundary are shown in
Figures 4.8 and 4.9. Values of thermal conductivity and density are considered to be
temperature independent and position dependent. For SUS3O4: ρm= 8166 kg/m3 and
km=12.04 W/m K; and for Si3N4 ρc= 2370 kg/m3, and kc=9.19 W/m K, where the subscripts
‘m’ and ‘c’ denote the metal and ceramic components, respectively. From the tables it can
be observed that increase in aspect ratio tends to increase the frequency of the plate and
maximum frequency is recorded for higher aspect ratio (a/b) =2.0. Difference between
frequency parameters having skew angle 0 and 15 is insignificant, while for 45 skew
angle noticeable difference in the frequency values may be observed.
159
14 24

0° 0°
15° 15°
22
30° 30°
45° 45°
12 20
18
10 16
14
8
12
10
6
0.5 1 2 Metal Ceramic 0.5 1 2 Metal
(a) Aspect ratio (a/b)=0.5 (b) Aspect ratio (a/b)=1.0
45
Skew angle
Skew angle

0° 70
40 0°
15°
15°
30°
30°
45° 60
35 45°
50
30
25 40
20 30
15 20
(c) Aspect ratio (a/b)=1.5 (d) Aspect ratio (a/b)=2.0
Figure 4.9 Influence of aspect ratio (a/b) on frequency parameter of clamped Si3N4/SUS3O4
plate for different skew angles (h=0.1m)
4.3.2.5 Influence of thickness ratio on non-dimensional frequency values

of Si3N4/SUS3O4 skew plate
Influence of thickness ratio (a/h) on frequency is demonstrated in Figure 4.10 for
Si3N4/SUS3O4 skew plate having clamped boundary condition. Linear variation of volume
fraction index (n=1.0) is assumed to accomplish the results. The top surface of the plate is
subjected to a temperature of 600K, while the bottom surface of the plate is subjected to
300K. For FGM plate with skew angle 45, free vibration frequency values are higher
compared to other skew angles. Also, increase in the thickness ratio of the plate increases
the frequency of the plate regardless of the values of skew angle and volume fraction
index.
160
Non-dimensional frequency parameter (  Skew angle

18 Skew angle 24
0°
0° 15°
15° 30°
16 30° 45°
45° 20
14
16
12
10
12

(a) Side-thickness ratio (a/h)=5.0 (b) Side-thickness ratio (a/h)=10
Figure 4.10 Influence of thickness ratio on frequency parameter of square clamped ZrO2/Ti-
6Al-4V plate for different skew angles
4.3.3 Free vibration frequency and forced vibration response of FGM

skew shell
In this sub section, free vibration frequency and forced vibration responses are presented
for FGM skew shell by assuming different material properties. Results are compared with
literature results for FGM shells having no skew boundary and new results are generated
for skew shell. All the results are presented in the following non-dimensional forms.
Frequency:   a 2 12m (1   2 ) Em h2
 a b  wEm h
Displacement: w  ,  
 2 2  qa
2
Time: t  t Em 2
a m
 a b   xx h
2
In-plane normal stresses:  xx  ,  
 2 2  qa
2
where the subscript ‘m’ represents the properties corresponding to metal shell.
4.3.3.1 Validation of free vibration frequency of simply supported

Si3N4/SUS3O4 shell
Free vibration frequency for simply supported Si3N4/SUS3O4 cylindrical shell is presented in
Table 4.24. First four mode shapes for different power law exponents n = 0.0, 0.2, 2.0, 10.0
and 1000 having geometric properties a/R=0.1 and a/h=10 are investigated. The literature
papers considered for validation purpose are: Neves et al. (2013b), who adopted HSDT in
161
conjunction with Carrera’s unified formulation (Carrera 2001; Carrera 2003; Carrera 2004)
and collocation radial basis technique (Ferreira 2003a; Ferreira 2003b; Ferreira et al. 2003;
Ferreira et al. 2006b); Pradyumna and Bandyobadhyay (2008), who presented free vibration
solution based on HSDT (Kant and Khare 1997) by using finite element method; and Yang
and Shen (2003), who carried out the vibration analysis based on HSDT (Reddy and Liu
1985) using semi analytical approach.
From Table 4.24, it may be concluded that the present results exhibit close range with the
above cited reference data for maximum number of cases. Maximum difference between the
free vibration frequency results is observed at first frequency mode with n = 0.0, 2.0 and
10.0. The probable reason may be due to the different higher order models involved in the
analysis. In case of Reddy’s higher order deformation theory (Reddy 1984b) used in present
research study, the unknowns present in the in-plane displacement fields are determined by
satisfying the condition of zero transverse shear stress at the top and bottom surface, which is
not in the case of displacement field proposed by Kant and Khare (1997). The different
methods proposed to extract the frequencies may be the influencing reason for the deviation.
Table 4.24: Validation study for free vibration of simply supported Si3N4/SUS3O4
cylindrical shell (a/h=10)
Mode
n References
1 2 3 4
Ceramic Present (12X12) 74.503 142.647 142.816 201.072
Yang and Shen (2003) 74.518 144.663 145.740 206.992
Pradyumna and Bandyopadhyay (2008) 72.961 138.555 138.555 195.536
Neves et al. (2013a) 74.263 141.677 141.848 199.156
0.2 Present (12X12) 60.834 116.431 116.587 164.113

Yang and Shen (2003) 57.479 117.717 112.531 159.855
Neves et al. (2013a) 60.006 114.378 114.549 160.735
2.0 Present (12X12) 40.585 77.356 77.451 108.754

Yang and Shen (2003) 40.750 78.817 79.407 112.457
Neves et al. (2013a) 40.525 76.972 77.081 107.948
10.0 Present (12X12) 35.090 66.709 66.787 93.629

Yang and Shen (2003) 35.852 69.075 69.609 98.386
Neves et al. (2013a) 35.166 66.648 66.732 93.335
Metal Present (12X12) 32.712 62.341 62.416 87.620

Yang and Shen (2003) 32.761 63.314 63.806 90.370
Neves et al. (2013a) 32.610 61.932 62.008 86.816
162
4.3.3.2 Validation of forced vibration response of Al/ZrO2 plate
Present (n=1), Praveen and Reddy (1998) (n=1),

5 Present (ceramic), Praveen and Reddy (1998) (ceramic),
Present (metal), Praveen and Reddy (1998) (metal)
0
Non-dimensional displacement (w)

-5
-10
-15
-20
-25
-30
-35
-40
0 2 4 6 8 10 12 14
Non-dimensional time (t)
Figure 4.11 Non-dimensional displacement of simply supported Al/ZrO2 square plate for
different values of volume fraction index
To perform validation of present results, simply supported Al/ZrO2 plate with a =b =0.2m
and h=0.01m is considered (Figure 4.11) in the present case. FGM plate is subjected to
uniformly distributed load of intensity 106 N/m2 in upward direction while a time step of
0.00001s is considered for the analysis. Displacement results are compared with those of
Praveen and Reddy (1998) based on FSDT (Reddy 1984b; Reddy 2004). The displacement
results are calculated for three different values of power law exponent n = 0.0, 1.0 and 1000,
and good agreement between the results is observed for all the values of volume fraction
indexes considered.
4.3.3.3 Influence of thickness ratio on free vibration response of Al/SiC
cylindrical shell
Tables 4.25 and 4.26 show the non-dimensional frequencies of square Al/SiC (R/a =5.0)
cylindrical skew shell having clamped and simply supported boundary conditions. It is
observed that, as the power law index rises, the frequency of the shell tends to reduce and
this is due to the low stiffness offered by the metal portion. Increasing trend of frequency
value with decreasing thickness of the shell is observed. Also, with the increase of skew
angle of the shell (i.e., beyond skew angle 30°), the frequency tends to increase at faster rate
(nearly about 1.5-1.7 times). Clamped skew shell establishes higher frequency compared to
simply supported shell, due to high stiffness.
163
Table 4.25: Non-dimensional frequency values of clamped Al/SiC square cylindrical shell
(R/a=5.0)
n
a/h  0 0.2 0.5 1.0 2.0 5.0 10 1000
15° 102.822 96.143 88.284 79.251 69.234 57.842 52.183 44.317
30° 101.371 94.744 86.956 78.070 68.282 56.850 51.134 43.376
5 45° 139.226 130.230 119.646 107.450 93.654 77.202 69.172 59.579
60° 216.383 202.808 186.871 168.348 146.744 120.123 107.135 92.582
15° 185.024 171.999 156.798 140.145 122.578 103.601 94.044 80.083

30° 171.945 160.384 147.136 132.399 116.610 98.979 89.621 75.080
10 45° 242.460 226.300 207.582 186.537 163.679 137.670 124.185 104.825
60° 395.899 370.347 340.463 306.286 267.921 222.554 199.645 170.403
15° 294.111 272.805 248.382 221.734 194.815 167.807 153.708 128.646

30° 299.849 280.400 258.372 233.763 206.533 175.248 158.265 131552
20 45° 417.148 389.657 358.343 323.420 285.271 242.099 218.796 181.977
60° 697.837 652.297 599.751 540.624 475.672 401.239 361.731 303.170
15° 534.163 495.534 451.386 403.477 355.577 309.121 284.618 234.978

30° 693.548 650.797 602.222 547.096 483.754 407.581 365.747 304.370
50 45° 934.133 875.746 809.441 734.455 648.985 547.264 491.578 408.749
60° 1523.040 1427.625 1318.461 1194.980 1055.548 891.609 802.190 667.230
15° 865.366 806.316 739.451 665.682 587.119 502.407 456.651 377.459

30° 1358.464 1275.945 1181.824 1074.461 950.124 799.230 716.050 596.066
100
45° 1813.851 1702.820 1576.456 1432.625 1266.382 1065.001 951.151 794.043
60° 2897.804 2720.454 2517.364 2286.156 2021.261 1704.125 1529.640 1272.777
Table 4.26: Non-dimensional frequency values of simply supported Al/SiC square

cylindrical shell (R/a=5.0)
n
a/h  Ceramic 0.2 0.5 1.0 2.0 5.0 10 Metal
15° 53.388 49.875 45.871 41.374 36.412 30.693 27.621 22.951
30° 82.586 76.946 70.400 62.981 54.933 45.898 41.425 35.264
5 45° 111.264 104.962 97.497 88.578 77.863 64.833 58.034 47.404
60° 150.283 143.009 134.086 122.837 108.356 90.098 80.542 64.179
15° 96.784 90.395 83.172 75.010 66.196 55.974 50.420 41.895

30° 138.752 128.777 117.872 105.667 92.876 79.113 71.918 60.337
10 45° 213.801 200.471 185.021 167.208 147.081 123.628 110.968 91.572
60° 294.951 278.867 260.227 238.182 211.245 176.641 156.895 125.992
15° 175.069 162.844 148.822 113.269 117.244 100.701 91.625 76.290

30° 235.901 219.941 201.892 181.905 160.377 136.613 123.879 103.166
20 45° 389.747 364.161 335.143 302.710 266.935 226.148 204.161 170.081
60° 583.545 549.221 510.231 465.263 411.876 344.735 306.428 249.947
15° 343.732 319.794 292.993 263.615 232.321 198.385 180.275 149.982

30° 510.505 476.861 438.863 396.507 349.894 297.010 268.362 223.337
50 45° 888.385 831.905 767.856 695.661 614.205 518.860 467.049 389.619
60° 1436.591 1354.542 1265.107 1132.626 996.447 835.543 746.387 612.483
15° 612.748 571.037 524.088 472.134 416.283 354.822 321.181 266.623

30° 946.338 885.230 816.317 739.081 652.493 551.682 496.762 413.288
100
45° 1706.310 1599.148 1473.810 1328.351 1170.123 995.443 896.831 745.520
60° 2745.516 2569.263 2369.666 2145.473 1893.306 1597.616 1436.002 1190.948
164
4.3.3.4 Influence of thickness ratio on free vibration response of Al/SiC
spherical shell
Table 4.27: Non-dimensional frequencies of clamped Al/SiC square spherical shell
n
a/h  Ceramic 0.2 0.5 1.0 2.0 5.0 10 Metal
15° 118.769 113.112 106.350 98.089 87.605 73.178 64.509 49.865
30° 134.209 127.982 120.436 111.092 99.095 82.544 72.829 56.552
5 45° 167.197 160.012 151.073 139.545 124.008 102.785 91.109 70.799
60° 231.432 221.269 207.937 189.837 165.221 136.131 122.957 98.608
15° 227.841 215.096 200.526 183.522 163.047 136.544 120.737 95.661

30° 249.303 236.188 220.802 202.578 180.478 151.577 134.413 107.090
10 45° 312.867 297.302 278.782 256.440 228.725 191.788 169.955 134.698
60° 444.292 422.317 394.756 358.670 315.584 263.484 236.125 192.118
15° 433.195 407.757 378.694 345.208 305.885 256.928 228.714 185.795

30° 481.982 453.905 421.762 384.733 341.196 286.645 255.106 206.980
20 45° 601.828 567.741 528.306 482.473 428.321 360.064 320.420 259.052
60° 841.972 790.077 729.640 660.709 583.011 491.312 441.658 368.184
15° 1055.636 990.086 916.386 833.110 737.080 619.584 552.284 453.260

30° 1170.697 1098.825 1017.694 925.462 818.106 685.900 611.378 502.225
50 45° 1437.485 1349.729 1250.066 1136.465 1004.953 844.381 753.608 618.200
60° 1942.545 1824.688 1689.009 1534.186 1356.430 1143.033 1025.769 852.656
100 15° 2118.873 1982.404 1837.416 1681.582 1464.397 1218.010 1090.172 903.134
30° 2320.251 2175.791 2013.102 1828.599 1614.298 1352.009 1205.708 994.916
45° 2787.506 2611.840 2413.901 2190.306 1933.477 1623.580 1451.427 1197.823
60° 3773.824 3546.135 3284.538 2985.526 2640.176 2222.549 1991.944 1654.534
Table 4.28: Non-dimensional frequencies of simply supported Al/SiC square spherical shell
a/h  n
Ceramic 0.2 0.5 1.0 2.0 5.0 10 Metal
5 15° 70.435 67.100 62.922 58.004 52.453 44.708 39.451 29.896
30° 91.233 87.173 82.240 76.273 68.905 58.169 51.061 38.501
45° 116.113 110.950 104.671 96.866 86.682 72.269 63.682 49.182
60° 151.826 145.167 136.969 126.514 112.545 93.545 82.937 64.716
10 15° 124.097 117.594 109.868 100.849 90.303 76.392 67.616 52.937

30° 168.884 160.016 149.738 137.711 123.203 103.687 91.410 71.268
45° 221.471 209.943 196.516 180.595 161.039 134.843 119.027 93.741
60° 296.372 281.168 263.429 242.203 215.706 180.290 159.414 126.318
20 15° 231.618 218.506 203.273 185.670 165.132 139.185 123.701 99.292

30° 325.842 306.822 285.331 260.683 231.492 194.069 171.782 137.498
45° 427.999 402.915 374.433 341.676 303.036 254.031 225.219 180.925
60° 585.131 551.643 513.447 469.126 416.014 347.946 308.485 249.207
50 15° 540.704 508.451 471.324 428.801 379.772 319.895 285.531 233.080

30° 796.679 747.061 691.631 628.853 555.625 464.637 412.593 336.357
45° 1031.472 966.306 893.666 811.658 716.652 599.803 533.327 435.128
60° 1432.093 1343.148 1243.084 1129.401 997.463 835.317 743.462 608.399
100 15° 1032.021 969.501 897.736 815.787 721.629 607.790 543.353 446.232
30° 1576.086 1475.905 1364.275 1238.315 1092.292 912.888 811.732 665.998
45° 2033.259 1902.225 1756.954 1593.624 1404.795 1173.654 1043.847 857.016
60° 2807.427 2627.588 2426.963 2200.913 1940.566 1624.170 1446.948 1189.335
165
Non-dimensional frequency values of square Al/SiC (R/a =5.0) spherical skew shell having
clamped and simply supported boundaries are presented in Tables 4.27 and Table 4.28. The
observations drawn for spherical shell are similar to that of cylindrical shells, except that
high magnitude of frequency is reported in case of spherical shell.
4.3.3.5 Influence of curvature ratio (R/a) on free vibration frequency of

Al/SiC cylindrical and spherical shell
240 240
200 200
Skew angle
15°
30° Skew angle
45° 160 15°
160
60° 30°
45°
60°
120 120
80 80
40 40
0 10 20 30 40 50 0 10 20 30 40 50
Radius to side ratio (R/a) Radius to side ratio (R/a)
(a) Clamped boundary condition (b) Simply supported boundary condition
Figure 4.12 Influence of R/a ratio on non-dimensional frequency of Al/SiC cylindrical skew
shell (n =1.0, a/h=10)
360 240
320 Skew angle

15° 200
280 30°
45° Skew angle
60° 15°
240 160
30°
45°
200 60°
120
160
80
120
80
40
0 10 20 30 40 50
0 10 20 30 40 50
Radius to side ratio R/a Radius to side ratio (R/a)
(a) Clamped boundary condition (b) Simply supported boundary condition
Figure 4.13 Influence of R/a ratio on non-dimensional frequency of Al/SiC spherical skew
shell (n =1.0, a/h=10)
166
This example refers to the free vibration study of square Al/SiC cylindrical and spherical
shell (a/h=10 and n =1.0) and having simply supported and clamped boundary conditions for
different curvature values (R/a). Several values of R/a ratio (0.2, 0.5, 5.0, 10.0, and 50.0) are
chosen to perform the study. The results showing influence of R/a ratio on frequency
parameter for cylindrical, spherical skew shell having simply supported and clamped
boundary condition are presented in Figures 4.12 and Figure 4.13, respectively (R=1/radius
of curvature). Up to a certain value of R/a (i.e., (R/a) =2.0), the free vibration frequency
values are decreasing, after which it converges to a constant for all R/a values assumed.
Cylindrical and spherical shell with clamped boundary shows higher frequency compared to
shell with simply supported boundary. In addition, spherical skew shell show lower free
vibration frequency values compared to cylindrical skew shell for different radius to side
(R/a) values considered in the example.
4.3.3.6 Free vibration frequency of simply supported and clamped Al/SiC

hypar shell
In this analysis, the term, c/a is used as an indicator to represent the twist curvature of hypar
shell. Influence of c/a ratio on frequency parameter for simply supported and clamped
boundary conditions having geometric properties a/h=10 and power law exponent (n) =1.0 is
presented in Table 4.29. When c/a ratio increases from 0.0 to 0.3, the frequency of the hypar
shell increases for all the skew angles () considered. As visualized in other cases also, shell
with clamped boundary shows more frequency than shell with simply supported boundary.
Table 4.29: Non-dimensional frequency values of square Al/SiC hypar shell (a/h=10.0)
Boundary c/a
condition
 0 0.05 0.1 0.15 0.2 0.25 0.3
Simply 15° 37.085 37.130 37.263 37.484 37.788 38.170 38.625
supported 30° 60.418 60.448 60.556 60.741 61.002 61.338 61.748
45° 100.105 100.136 100.238 100.408 100.647 100.955 101.331
60° 177.715 177.745 177.806 177.897 178.018 178.170 178.352
Clamped 15° 56.200 56.254 56.417 56.687 57.063 57.542 58.121

30° 78.036 78.072 78.174 78.342 78.576 78.874 79.236
45° 118.182 118.214 118.306 118.458 118.670 118.941 119.271
60° 199.356 199.376 199.424 199.500 199.603 199.735 199.894
4.3.3.7 Forced vibration response of Si3N4/SUS3O4 skew shell

A FGM Si3N4/SUS3O4 cylindrical skew shell is considered in these examples to study the
effects of different parameters such as skew angle (), volume fraction index (n), shell
geometry (cylindrical and spherical) and aspect ratio (b/a). The results are presented in the
167
form of Figures. Simply supported boundary condition is considered in all the numerical
problems and displacements are presented in non-dimensional form. In the first example, to
study the consequence of change of skew angle on the central displacement, FGM
cylindrical shell having a/h=10.0 is analysed and the results obtained are shown in Figure
4.14. To perform the analysis, value of the skew angle is varied from 15° to 60° and linear
variation of n (n=1.0) is considered. It is observed that the cylindrical shell having skew
value 30° shows the maximum displacement; and the minimum displacement is observed for
skew angle 60°. Therefore it can be concluded that, an increase in skew angle contributes
more stiffness to the shell thus recording minimum displacement at the center of the shell.
Figure 4.15 presents the effect of aspect ratio (b/a) on central displacement of
cylindrical shell having skew angle () =15°. Four different values of aspect ratio (b/a= 0.5,
1.0, 2.0 and 5.0) are chosen to perform the study. Smaller value of aspect ratio (b/a=0.5)
gives maximum displacement while the minimum value of displacement is observed for
higher value of b/a=5.0. In Figure 4.16 central displacement response of skew shell ( =15°)
having cylindrical and spherical shell forms are considered. As observed in previous
examples, spherical shell report lower deflection compared to cylindrical shell, thus showing
more stiffness contributing property. In the next study, volume fraction index (n) is varied
from ceramic (n=0) to metal segment (n=∞), to study its influence on transient response of
cylindrical skew shell as in Figure 4.17.
100 Skew angle

0°
80 15°
30°
60 45°
60°
40
20
-20
-40
-60
-80
0 3 6 9 12 15
Figure 4.14 Influence of skew angle on the dynamic response of Si3N4/SUS3O4 cylindrical
skew shell (n =1.0, a/h=10)
168
Aspect ratio
1.6 b/a=0.5
b/a=1.0
1.4 b/a=2.0

b/a=5.0
1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
0 2 4 6 8 10 12 14 16
Figure 4.15 Influence of aspect ratio (b/a) on transient response of Si3N4/SUS3O4 cylindrical
skew shell (=15°, n=1.0, a/h=10)
0.30
Cylindrical
Spherical
0.25
0.20
0.15
0.10
0.05
0.00
-0.05
-0.10
0 3 6 9 12 15
Figure 4.16 Influence of shell geometry on the dynamic response of Si3N4/SUS3O4 skew shell
( =15°, n =1.0, a/h=10)
Cylindrical metal shell records maximum displacement, followed by composite shell and
pure ceramic shell (n=0.0). Dominance of stiffness effect by ceramic shell is the possible
cause for this type of response. In Figure 4.18, variation of in-plane normal stresses over a
period of time for cylindrical shell having skew angles from 0° to 60° is studied. It is
observed that the cylindrical shell with skew angle 30° gives maximum normal stress
compared to other cases of skew shells.
169
n=0
n=1
0.3 n=5

n=10
n=1000
0.2
0.1
0.0
-0.1
0 3 6 9 12 15
Figure 4.17 Influence of volume fraction index (n) on the dynamic response of Si3N4/SUS3O4
cylindrical skew shell (=15°, a/h=10)
0.008
0.006
Non-dimensional axial stresses (xx)
0.004
0.002
0.000
-0.002
-0.004
Skew angle
15°
-0.006
30°
45°
-0.008 60°
0 2 4 6 8 10 12 14 16
Figure 4.18 Non-dimensional in-plane normal stresses of Si3N4/SUS3O4 cylindrical

skew shell (n =1.0, a/h =100)
4.4 BUCKLING ANALYSIS OF FGM SKEW PLATE SUBJECTED TO IN-PLANE

MECHANICAL LOADING AND THERMAL FIELDS
In this section, buckling analysis of FGM skew plates is performed under various
combinations of in-plane mechanical loads and temperature effects. For thermal loading,
temperature dependent material properties are taken into account. In addition influence of
170
several other parameters such as aspect ratio, thickness ratio, skew angle and volume
fraction index are incorporated in the study. The non-dimensional buckling load parameter
is represented by the following expression.
N cr b 2 Ec h3
Critical buckling load parameter: N cr  2 where c
D 
 Dc 12(1  2 )
4.4.1 Validation of the results of buckling analysis of Al/Al 2O3 skew plate
subjected to uni-axial load
Table 4.30: Convergence and comparison study of critical buckling load for simply
supported Al/Al2O3 square skew plate subjected to uni-axial in-plane load ( h=0.01m)
n
ψ References
0 0.5 1.0 2.0 5.0 10.0
1
0° Present (4x4) 4.0006 1.9866 1.5878 1.3424 1.1372 0.9804
Present (6x6) 3.9979 1.9852 1.5867 1.3414 1.1364 0.9798
Present (8x8) 3.9974 1.9850 1.5865 1.3413 1.1363 0.9796
Present (12x12) 3.9973 1.9849 1.5865 1.3412 1.1362 0.9796
Present (16x16) 3.9972 1.9849 1.5865 1.3412 1.1362 0.9796
Present (20x20) 3.9972 1.9849 1.5865 1.3412 1.1362 0.9796
Ganapathi et al. (2006) 4.0010 2.1587 1.7956 1.5256 1.2559 1.0797
Wang (1997) 4.0000 - - - -
% of difference 0.09 8.05 58.65 0 0 0
15° Present (20x20) 4.3964 2.1834 1.7451 1.4752 1.2496 1.0773

Ganapathi et al. (2006) 4.3946 2.3713 1.9716 1.6752 1.3800 1.1868
Wang et al. (1997) 4.3937 - - - - -
% of difference 0.04 7.92 11.49 11.94 9.44 9.23
30° Present (20x20) 5.9740 2.9704 2.3736 2.0048 1.6963 1.4624

Ganapathi et al. (2006) 5.8966 3.1935 2.6496 2.2515 1.8607 1.6032
Wang et al. (1997) 5.8963 - - - - -
% of difference 1.31 6.99 10.42 10.95 8.84 8.78
45° Present (20x20) 10.7449 5.3644 4.2835 3.6071 3.0396 2.6199

Ganapathi et al. (2006) 10.1031 5.5139 4.5445 3.8625 3.2234 2.7964
Wang et al. (1997) 10.1026 - - - - -
% of difference 6.35 2.72 5.74 6.61 5.70 1.62
In the first case, buckling problem of Al/Al2O3 skew plate is studied under uni-axial in-
plane loads. Buckling load results for different skew angles are shown in Table 4.30 for
different mesh sizes to show the convergence. Percentage difference between the present
buckling load results and the results obtained by Ganapathi et al. (2006) is also shown in
Table 4.30. The buckling load results obtained are found to have small deviation from the
results reported in the literature due to the implementation of present TSDT which assumes
parabolic variation of shear strain while in case of buckling load results by Ganapathi et al.
(2006), FSDT model is incorporated. It may be noted that for higher values of volume
171
fraction index (n) the critical buckling load of the FGM plate decreases. This observation is
due to the fact that, the stiffness of the plate becomes lesser with the higher volume
fraction index. It is also manifested that as the skew angle of the plate increases the critical
buckling parameter also increases for all the values of volume fraction index.
To show the influence of homogenization approach on buckling parameter, second
validation example is performed for a simply supported Al/Al2O3 skew plate under in-
plane axial load. In Table 4.31, critical buckling load values obtained from the present
work are compared with the results of Ganapathi et al. (2006), which are based on FSDT.
Percentage difference between the present buckling load results and the results obtained by
Ganapathi et al. (2006) is also shown in Table 4.31. In this analysis, thickness ratio (a/h) is
taken as 100, volume fraction index (n) is varied from 0 to 2.0, and skew angle (ψ) is
varied from 0° to 30°. From Table 4.31, it can be visualized that the critical buckling loads
obtained by the present TSDT theory are lower than FSDT results. This is due to the fact
that, in the present TSDT, parabolic variation of shear stress is implemented. Differences
in results of buckling load parameter based on Voigt rule of mixture and Mori-Tanaka
scheme can also be observed in Table 4.31. It can be observed that, Voigt rule of mixture
(VRM) predicts higher values of buckling load compared to the Mori-Tanaka scheme
(MTS). The results obtained by Mori-Tanaka scheme (MTS) give true representation of
critical buckling load as compared to Voigt rule of mixture (VRM), because the
interactions among the elastic parameters are considered in case of former method. Also, it
is observed that the critical buckling load of FGM skew plate decreases as the volume
fraction index (n) increases, as it correspond to the low stiffness of the metal portion. The
other outcome of Table 4.31 is that, as the skew angle increases from 0°-30°, the buckling
load of the FGM skew plate increases irrespective of homogenization method and volume
fraction index.
Table 4.31: Convergence and comparison of results of critical buckling load for simply
supported Al/Al2O3 square skew plate subjected to uni-axial in-plane load ( h=0.01m)
n
Ψ References 0 0.5 1 2
VRM MTS VRM MTS VRM MTS VRM MTS
0° Present 3.9939 3.9940 2.6066 1.9833 1.9908 1.5851 1.5534 1.3402
Ganapathi et al. (2006) 4.0010 - 2.7061 2.1587 2.2054 1.7956 - 1.5256
% of difference 0.17 - 3.68 8.12 47.32 11.72 - 12.15
15° Present 4.0985 4.0986 2.6752 2.0355 2.0434 1.6269 1.5943 1.3753
Ganapathi et al. (2006) 4.3946 - 2.9795 2.3713 2.4277 1.9716 - 1.6752
% of difference 6.74 - 30.43 14.16 15.83 17.48 - 17.90
30° Present 4.4768 4.4769 2.9251 2.2260 2.2356 1.7788 1.7440 1.5204
(VRM-Voigt rule of mixture; MTS-Mori-Tanaka scheme)
172
4.4.2 Validation of the results of buckling analysis of isotropic and
Al/Al2O3 FGM skew plate subjected to thermal loading
To validate the present formulation for calculating the critical buckling temperature, first
the problem of an isotropic plate solved by Gossard et al. (1952) and Singh et al. (2001) is
solved and the results obtained are presented in Table 4.32. Also, percentage difference
between the present buckling load results and the results obtained by Gossard et al. (1952)
for isotropic plate is shown in Table 4.32. Different geometric and material parameters
considered are: thickness ratio (a/h) = 100; Poisson’s ratio () = 0.3 and co-efficient of
linear expansion (α) = 2.0 x10-6/°C. Two types of boundary conditions namely, simply
supported and clamped are considered for the purpose of comparison. Good agreement
among the results may be observed in the present study. Furthermore, it is found that the
20 x 20 mesh division provides sufficiently good convergence.
Table 4.32: Non-dimensional critical buckling load of square isotropic plate subjected to uni-
axial and bi-axial in-plane loads (h=0.01m)
References Simply supported Clamped
Present (4x4) 123.67 339.55
Present (6x6) 76.76 225.58
Present (8x8) 67.59 188.45
Present (12x12 ) 64.09 171.84
Present (16x16) 63.49 168.83
Present (20x20) 63.33 167.99
Gossard et al. (1952) 63.27 168.71
% of difference 0.09 0.42
Singha et al. (2001) 63.26 167.85
To perform the validation for FGM plate, Al/Al2O3 plate with thickness ratio a/h =
50 and 100 is considered under uniform temperature rise. The buckling results obtained are
compared with the results of Zhao et al. (2009b) in Table 4.33 for different boundary
conditions (simply supported and clamped). It is experienced that the buckling temperature
is more for lower value of thickness ratio (a/h = 50) and also with increasing constraints at
the boundary of the plate increases the critical buckling temperature. Also, it is noticed that
Mori-Tanaka scheme (MTS) predicts higher values of critical buckling temperature to
those of the Voigt rule of mixture (VRM). The role played by the parameter n on critical
buckling temperature is same as mentioned in the preceding explanation corresponding to
Table 4.31.
173
Table 4.33: Non-dimensional critical buckling temperature of Al/Al2O3 square plate under
uniform temperature rise
n
Boundary
a/h References 0 1 3
conditions
VRM MTS VRM MTS VRM MTS
50 Present 68.199 68.197 36.712 45.385 36.831 38.516
Zhao et al. ( 2009b) 67.915 - 31.185 - 29.608 -
% of error 1.49 - 17.72 - 24.39 -
SSSS
100 Present 17.09 17.09 9.1969 11.373 9.2317 9.656
Zhao et al. (2009b) 17.361 - 7.944 - 7.423 -
% of error 1.56 - 1.2529 - 24.37 -
50 Present 179.848 179.844 83.648 99.266 73.701 87.368

Zhao et al. (2009b) 179.817 - 82.357 - 74.591 -
% of error 0.02 - 1.57 - 1.19 -
CCCC
100 Present 45.252 45.250 21.029 24.974 18.547 22.012
Zhao et al. (2009b) 44.171 - 20.771 - 19.150 -
% of error 2.45 - 1.24 - 3.15 -
4.4.3 Buckling analysis for FGM skew plate subjected to uni-axial and bi-
axial in plane loads
Table 4.34: Non-dimensional critical buckling load for simply supported Al/ZrO2 square
skew plate subjected to uni-axial and bi-axial in-plane loads ( h=0.1 m)
Boundary Loading n ψ
conditions 0° 15° 30° 45° 60°
Uni-axial 0 37.2854 38.0313 40.2856 43.1959 40.5265
1.0 44.5339 46.2560 52.2829 65.6238 85.3525
2.0 11.5269 11.7496 12.4118 13.1913 12.0983
SSSS
Bi-axial 0 18.6427 18.3287 17.4513 16.1404 14.1283
1.0 6.9451 6.8278 6.5002 6.0104 5.2594
2.0 5.7635 5.6636 5.3827 4.9559 4.2886
Uni-axial 0 80.3226 79.1715 74.9148 65.6128 49.7563

1.0 29.9653 29.5389 27.9621 24.5275 18.6820
2.0 24.4143 24.0225 22.6077 19.6198 14.6825
CCCC
Bi-axial 0 44.0384 43.2674 40.8244 36.2002 28.0838
1.0 16.4003 16.1124 15.2008 13.4788 10.4697
2.0 13.4386 13.1895 12.4004 10.9125 8.3406
Uni-axial 0 54.1479 53.7452 53.5695 51.3184 42.7513

1.0 20.1889 20.0401 19.9740 19.1325 15.9672
2.0 16.6113 16.4700 16.3553 15.5330 12.7311
SCSC
Bi-axial 0 28.7198 27.8933 25.9068 22.5954 17.5711
1.0 10.7012 10.3938 9.6535 8.4181 6.5441
2.0 8.8256 8.5656 7.9371 6.8879 5.3016
In this sub section, results of buckling analyses are shown in Table 4.34 for simply
supported Al/ZrO2 skew plate under uni-axial and bi-axial compression. It may be
visualized that, more buckling load is required in the case of uni-axial loading than bi-axial
174
loading to buckle the plate. Also, increasing constrains on the boundary of the plate
increases the buckling load parameter. Further, if the value of volume fraction index
increases, the critical buckling load of the plate reduces. This is due to the lower stiffness
value of the metal component at the bottom of the plate.
Buckling load parameters for different volume fraction index (n = 0, 0.5, 1.0 and
2.0) and skew angles (ψ = 0°, 15°, 30°, 45° and 60°) with thickness ratio a/h =10 is
presented in Table 4.35. Both uni-axial and bi-axial loading conditions are considered and
the properties are estimated by Mori-Tanaka scheme (MTS) and Voigt rule of mixture
(VRM).
Table 4.35: Non-dimensional critical buckling load of clamped Al/Al2O3 plate under in-plane
loading for different skew angles (Ψ)
n
ψ Ncr 0 0.5 1.0 2
VRM MTS VRM MTS VRM MTS VRM MTS
0° uni-axial 8.0303 8.0324 5.3362 4.0399 4.0892 3.1907 3.1467 2.6287
bi-axial 4.7039 4.4039 2.9132 2.2066 2.2293 1.7464 1.7194 1.4467
15° uni-axial 7.9172 7.9173 5.2665 39858 4.0368 3.1454 3.1034 2.5867
bi-axial 4.3267 4.3268 2.8640 2.1689 2.1920 1.7158 1.6896 1.4199
30° uni-axial 7.4915 7.4916 5.0038 3.7831 3.8388 2.9776 2.9423 2.4351
bi-axial 4.0825 4.0825 2.7084 2.0497 2.0736 1.6189 1.5954 1.3353
45° uni-axial 6.5613 6.5614 4.4179 3.3343 3.3957 2.6121 2.5887 2.1143
bi-axial 3.6200 3.6208 2.4140 1.8241 1.8498 1.4357 1.4175 1.1756
60° uni-axial 4.9756 4.9756 3.3970 2.5580 2.6217 1.9893 1.9821 1.5831
bi-axial 2.8084 2.8084 1.8945 1.4274 1.4552 1.1154 1.1059 0.8992
Values of critical buckling load are more in case of uni-axial compression
compared to critical buckling load under bi-axial compression. For the plate corresponds to
full ceramic component (n=0) both Voigt rule of mixture (VRM) and Mori-Tanaka scheme
(MTS) show similar trend. However, for higher values of volume fraction index (n= 0.5,
1.0 and 2.0), Voigt rule of mixture (VRM) gives higher values of buckling load for all
skew angles. From Table 4.35, it is inferred that for skew angle (Ψ = 0ο to 45ο), values of
buckling load are higher and for further increase in skew angle (Ψ = 60ο), lower values of
critical buckling load are observed.

175
4.4.4 Influence of aspect ratio (a/b) on critical buckling temperature of
simply supported Al/Al2O3 skew plate subjected to uniform temperature
rise
Variations of critical buckling temperature for various aspect ratio (a/b= 1,2,3,4 and 5) and
skew angles (ψ=0°,15°,30°, 45°and 60°) using Voigt rule of mixture (VRM) and Mori-
Tanaka scheme (MTS) are depicted in Figure 4.19. Higher value of buckling temperature
is observed for plate corresponds to aspect ratio (a/b) = 5 in both types of approaches.
Further, increase in skew angle of the plate shows higher buckling load of the plate under
thermal condition. For different values of skew angles, initially the buckling load is
observed to increase slowly but for skew angle more than 45° it changes quire rapidly.
14000
Aspect ratio
a/b=1 Aspect ratio
12000 a/b=2 12000 a/b=1
Critical buckling temperature (
a/b=2
Critical buckling temperature (
a/b=3
a/b=4 a/b=3
10000 10000 a/b=4
a/b=5
a/b=5
8000 8000
6000 6000
4000
4000
2000
2000
0
0° 15° 30 ° 45° 60° 0° 15° 30 ° 45° 60°
Skew Angle Skew Angle
(a) Voigt-rule of mixture (b) Mori-Tanaka scheme
Figure.4.19. Influence of aspect ratio (a/b) on non-dimensional buckling temperature of

simply supported Al/Al203 plate under uniform temperature rise (a/h=10 and n=1)
4.4.5 Critical buckling temperature of simply supported Si3N4/ SUS3O4

skew plate having thermal dependent properties and subjected to
Effect of temperature-dependent material properties on the critical buckling temperature is
investigated by considering simply supported Si3N4 (ceramic) / SUS3O4 (metal) skew plate.
Thermal dependent material properties used for Si3N4/ SUS3O4 skew plate are listed in
Table 4.17. VRM is used to calculate the effective properties which are dependent on both
temperature and position. In this example, the Young’s modulus (E) and co-efficient of
thermal expansion (α) are considered to be temperature-dependent as per Table 4.36.
176
Values of mass density (ρ) and thermal conductivity (k) of the constituents are temperature
independent and assumed as: ρc =2370 kg/m3; ρm = 8166 kg/m3; kc= 9.19 W / m.K and km
= 12.04 W /m.K (where subscripts ‘c’ and ’m’ denotes the ceramic and metal plates,
respectively). Thickness ratio (a/h =10) is considered in this example and the critical
buckling temperature is evaluated for linear variation of ceramic and metal (n=1). In Table
4.36, variation of critical buckling temperature is reported for different temperature
differences (∆T) varying from 200K to 1000K. From Table 4.36, it can be concluded that
the buckling load of the plate reduces as the temperature difference increases. Weak
Young’s modulus (E) of the material at high temperature is the reason for such
observation. Also, with increase in skew angle, increase in thermal load carrying capacity
of the plate is observed irrespective of the value of n.
Table 4.36: Critical buckling temperature (ΔTcr) of simply supported Si3N4/ SUS3O4 skew
plate for various skew angles (a/h =10 and n = 1.0)
Temperature difference (K)
ψ
200 400 600 800 1000
0° 9.2604 4.6301 3.0867 2.3151 1.8520
15° 9.7124 4.8562 3.2374 2.4281 1.9425
30° 11.512 5.7562 3.8375 2.8782 2.3025
45° 15.973 7.9869 5.3246 3.9935 3.1947
60° 27.504 13.7520 9.1680 6.8760 5.5008
4.5 STATIC, FREE VIBRATION AND BUCKLING ANALYSES OF FGM PLATE

DESCRIBED BY FOUR-PARAMETER DEPENDENT POWER LAW
DISTRIBUTION
As briefed in previous chapters, conventional FGM represent the ceramic at top and metal
at bottom with functional graded properties in the thickness direction of geometry. To
obtain the
Table 4.37: Different profiles and values of material gradient parameters
Types Parameters Profiles Distribution
FGM1 a1=1.0, b1=c1=0 Classical Metal at top and ceramic at bottom
FGM2 a1=1.0, b1=1.0, c1=2.0 Symmetric Ceramic at top and bottom
FGM3 a1=1.0, b1=1.0, c1=4.0 Asymmetric Ceramic at top and bottom
FGM4 a1=1.0, b1=0.5, c1=2.0 Asymmetric Ceramic at bottom/top and mixture of metal and ceramic at
top/bottom
top/bottom
top/bottom
177
4.5.1 Static analysis of Al/ZrO2 plate subjected to mechanical loading
considering different FGM profiles
0.016
0.0024
0.014
0.0020
Displacement (w)
FGM1
0.012
Displacement (w)
FGM2
FGM1
FGM3
FGM2
FGM4 0.0016 FGM3
0.010 FGM5 FGM4
FGM5
0.008 0.0012
0.006
0 20 40 60 80 100 0 20 40 60 80 100
(a) Clamped-free boundary condition 0.0072 (b) Clamped boundary condition
0.0032
0.0066
0.0028
0.0060
Displacement (w)
FGM1
Displacement (w)
0.0054
0.0024 FGM2 FGM1
FGM3 FGM2
0.0048
FGM4 FGM3
FGM5 FGM4
0.0020
0.0042 FGM5
0.0036
0.0016
0.0030
0 20 40 60 80 100 0 20 40 60 80 100
(c) Simply supported-clamped boundary condition (d) Simply supported boundary condition
Figure 4.20 Non-dimensional deflections of Al/ZrO2 plate having different boundary

conditions based on four-parameter power law function (a/h=10)
configuration other than conventional one, a four-parameter dependent power law is
proposed in the literature (Tornabene 2009). In this section, a four-parameter power law
distribution represented by equation (3.5.1) in Chapter 3 is used to perform the static,
vibration and buckling analyses of FGM plate. The selection range of various material
properties to achieve the different kinds of FGM plates is furnished in Table 4.37. To
exhibit the deflection results based on four-parameter FGM distribution, different profiles
namely classical, symmetric and asymmetric profiles are assumed in this section.
178
160 n=0 160 n=0
n=0.5 n=0.5
n=1 140 n=1
Dimensioned displacement (w)

n=5 n=5
120 n=10 120 n=10
Displacement (w)
n=20 n=20
100 n=50
n=50
n=100 n=100
80
80
60
40
40
20
0
0
0 20 40 60 80 100 0 20 40 60 80 100
Thickness ratio (a/h) Thickness ratio (a/h)
(a) FGM1 (b) FGM2
160 n=0
n=0.5
n=1
n=5
120 n=10
n=20
Displacement (w)
n=50
n=100
80
40
0 20 40 60 80 100
Thickness ratio (a/h)
(c) FGM3
Figure 4.21 Influence of thickness ratio (a/h) on non-dimensional deflection of Al/ZrO2 plate
for different volume fraction index based on four-parameter power law function
Six kinds of FGM profiles under different boundary conditions and thickness value (h)
0.1m are assumed. Of different support conditions, FGM plate with all the edges clamped
records minimum deflection in the group with volume fraction index (Figure 4.20). Also,
different FGM models exhibit same magnitude of deflection for lower values of material
gradient index (n= 0, 0.5, 1.0 and 5.0). Further increase in the value of n causes the
deflection to rise gradually and jumps into stable mode beyond n =50. The reason this may
be when n elevates the gradation property diminishes slowly i.e., plate contains high
percentage of isotropic material.
179
0.024
0.024 n=0
n=0 n=0.5
n=0.5 n=1
0.020
0.020 n=1 n=5
n=5 n=10
n=10
Displacement (w)
0.016 n=20
0.016 n=20 n=50
Displacement (w)
n=50
n=100
n=100 0.012
0.012
0.008 0.008
0.004 0.004
0.000 0.000
0 1 2 3 4 5 0 1 2 3 4 5
Aspect ratio (a/b) Aspect ratio (a/b)
(a) FGM1 (b) FGM2
0.024
n=0
n=0.5
0.020 n=1
n=5
n=10
Displacement (w)
0.016 n=20
n=50
n=100
0.012
0.008
0.004
0.000
0 1 2 3 4 5
Aspect ratio (a/b)
(c) FGM3
Figure 4.22 Influence of aspect ratio (a/b) on non-dimensional deflection of Al/ZrO2 plate for
different volume fraction index based on four-parameter power law function (a/h=10)
In addition, among the various profiles shown, symmetric profiles show better
performance as they give lower deflections. Except FGM2 model, all the models shows no
significant divergence in the behavior and this trend is found to be marked when n ranges
from 5 to 50. To sketch the influence of side-thickness (a/h) ratio on deflection, three kinds
of profiles (FGM1, FGM2 and FGM3) are assumed as shown in Figure 4.21 with several
ranges of volume fraction index (n= 0 to 100). When the thickness range falls under thick
category, there will not be any significant improvement in deflection parameter. Variation
of deflection is observed to be drastic for thin models and this fact is identical for all FGM
models under consideration. Symmetric profiles exhibit lower deflection followed by
asymmetric and classical profiles, once again.
180
In addition, parametric study was performed to study the pattern of deflection for
different aspect ratios (a/b) by incorporating different material gradient index (Figure
4.22).
0.6 0.6
n=0 n=0
0.5
n=0.5 0.5
n=0.5
0.4 n=1 0.4 n=1
n=5 n=5

0.3 0.3
n=10 n=10
0.2 0.2
0.1 0.1
0.0 0.0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
-4.00E+007 -2.00E+007 0.00E+000 2.00E+007 4.00E+007 -4.00E+007 -2.00E+007 0.00E+000 2.00E+007 4.00E+007
In-plane normal stress (xx) In-plane normal stress (xx)
0.6
(a) FGM1 (b) FGM2
n=0 0.6
0.5 n=0
n=0.5 0.5 n=0.5
0.4 n=1 n=1
0.4
n=5 n=5
0.3
n=10 0.3 n=10
0.2
0.2
0.1
0.1
0.0 0.0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
-4.00E+007 -2.00E+007 0.00E+000 2.00E+007 4.00E+007 -4.00E+007 -2.00E+007 0.00E+000 2.00E+007 4.00E+007
(c) FGM3 (d) FGM4
0.6
n=0
0.5 n=0.5
0.4 n=1
n=5
0.3 n=10
0.2
0.1
0.0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-4.00E+007 -2.00E+007 0.00E+000 2.00E+007 4.00E+007
In-plane normal stress (xx)
(e) FGM5
Figure 4.23 Through-the-thickness variation of in-plane normal stress (σxx) of Al/ZrO2 plate
for different volume fraction index based on four-parameter power law function (a/h=10)
181
0.6 0.6
n=0 n=0
0.5 0.5 n=0.5
n=0.5
0.4 n=1 0.4 n=1
n=5 n=5
0.3 0.3 n=10
n=10

0.2 0.2
0.1 0.1
0.0 0.0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
-4.00E+007 -2.00E+007 0.00E+000 2.00E+007 4.00E+007-4.00E+007 -2.00E+007 0.00E+000 2.00E+007 4.00E+007
In-plane normal stress (yy) In-plane normal stress (yy)
(a) FGM1 0.6 (b) FGM2
0.6 n=0
n=0
0.5 n=0.5
0.5 n=0.5
n=1 0.4 n=1
0.4
n=5 n=5
0.3 n=10
0.3 n=10
0.2 0.2
0.1 0.1
0.0 0.0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
-4.00E+007 -2.00E+007 0.00E+000 2.00E+007 4.00E+007 -4.00E+007 -2.00E+007 0.00E+000 2.00E+007 4.00E+007
(c) FGM3 (d) FGM4
0.6
n=0
0.5 n=0.5
0.4 n=1
n=5
0.3
n=10
0.2
0.1
0.0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-4.00E+007 -2.00E+007 0.00E+000 2.00E+007 4.00E+007
In-plane normal stress (yy)
(e) FGM5
Figure 4.24 Through-the-thickness variation of in-plane normal stress (σyy) of Al/ZrO2 plate
The parameter b1 is kept constant, while a1 is varied from 0.5 to 5.0. The curves
representing various volume fraction index could not show much disparity with respect to
182
deflection response for lower aspect ratio, and show gradual increasing trend as aspect
ratio becomes greater than 0.5. Asymptotic response was manifested when the length of
the plate turn into three times its width (i.e., a=3b).
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3

0.2 0.2
n=0
0.1 n=0.5 0.1
n=1
0.0 0.0
n=5
-0.1 n=10 -0.1 n=0
n=0.5
-0.2 -0.2
n=1
-0.3 -0.3 n=5
-0.4
n=10
-0.4
-0.5 -0.5
-0.6 -0.6
-2.00E+007 0.00E+000 2.00E+007 -2.00E+007 0.00E+000 2.00E+007
0.6 In-plane shear stress (xy) 0.6 In-plane shear stress (xy)
0.5 (a) FGM1 (b) FGM2
0.5
0.4
0.4
0.3 0.3
0.2 0.2
n=0 n=0
0.1 n=0.5 0.1 n=0.5
n=1 n=1
0.0 0.0
n=5 n=5
-0.1 n=10 -0.1 n=10
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
-2.00E+007 0.00E+000 2.00E+007 -2.00E+007 0.00E+000 2.00E+007
In-plane shear stress (xy) In-plane shear stress (xy)
(c) FGM3 0.6
(d) FGM4
0.5
0.4
0.3
0.2
0.1 n=0
n=0.5
0.0
n=1
-0.1 n=5
n=10
-0.2
-0.3
-0.4
-0.5
-0.6
-2.00E+007 0.00E+000 2.00E+007
In-plane shear stress (xy)
(e) FGM5
Figure 4.25 Through-the-thickness variation of in-plane shear stress (𝛕xy) of Al/ZrO2 plate
183
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3

0.2 0.2 n=0

n=0 n=0.5
0.1 0.1
n=0.5 n=1
0.0 n=1 0.0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
0 750000 1500000 2250000 3000000 3750000 0 750000 1500000 2250000 3000000 3750000
Transverse shear stress (yz) Transverse shear stress (yz)
(a) FGM1 (b) FGM2
0.6
0.6
0.5
0.5
0.4
0.4
0.3
0.3
0.2 n=0
0.2
n=0 0.1 n=0.5
0.1 n=0.5 n=1
n=1 0.0
0.0
-0.1
-0.1
-0.2
-0.2
-0.3
-0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
0 750000 1500000 2250000 3000000 3750000 0 800000 1600000 2400000 3200000
Transverse shear stress (yz) Transverse shear stress (yz)
(c) FGM3 (d) FGM4
0.6
0.5
0.4
0.3
0.2 n=0
0.1 n=0.5
n=1
0.0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
0 750000 1500000 2250000 3000000 3750000
Transverse shear stress (yz)
(e) FGM5
Figure 4.26 Through-the-thickness variation of transverse shear stress (𝛕yz) of Al/ZrO2 plate
Except FGM1 model, plate containing lower percentage of volume fraction of
ceramic/metal, does not show any marked response for different values of a/b. In all the
184
three cases shown in Figure 4.22, FGM model with high value of n (n=100) shows
maximum deflection, due to less stiffening effect offered by the plate.
In Figure 4.23, variation of in-plane stress (σxx) is plotted through the thickness for
different kinds of FGM profiles. Linear variation of stress is observed through the
thickness, and this trend changes for n> 5.0. Non-linear variation of compressive and
tensile stresses was noticed for n=10.0, but stresses are lower in magnitude. This implies
the presence of larger volume fractions of ceramic and metal constituents. However, the
volume fraction of particular constituent along the thickness direction depends on type of
FGM model under consideration. Among different models, FGM5 model shows maximum
tensile and compressive stresses at top and bottom, respectively, for different values of
volume fraction gradient indexes.
In plot of axial stress variation (σyy) through the thickness (Figure 4.24), the nature
of stresses developed is analogous to those observed in Figure 4.23. High magnitude of
compressive stresses at bottom zone is observed for isotropic plate. At top segment of
plate, stresses exhibited by isotropic and graded plates are almost similar and considerable
deviation is noticed when n=10. In addition, graded plates record minimum stresses
(compressive) compared to isotropic plates for all FGM models. The conclusion drawn for
in-plane axial stresses in y direction holds true for in- in-plane shear stresses (𝛕xy) (Figure
4.25), but the difference is observed in view of nature of stresses developed at top and
bottom of the plate. To represent the accuracy of the present formulation in predicting the
transverse shear mode, variation of transverse shear stress (τyz) through the thickness is
depicted in Figure 4.26. Stresses are shown for three ranges of material gradient index
(n=0, 0.5 and 1.0) and symmetric profile shows minimum stresses (compression and
tension) like other cases.
4.5.2 Free vibration frequency of Al/ZrO2 plate by considering different

FGM profiles
4.5.2.1 Influence of boundary conditions and thickness on free vibration
analysis of four-parameter Al/ZrO2 plate
In Tables 4.38 and 4.39, first six natural frequencies of simply supported thin (h=0.001 m)
and moderately thick (h=0.1m) Al/ZrO2 FGM plate are furnished. Three types of power
law profiles are considered for each case viz., classic, symmetric and asymmetric. To
185
perform the study, values of volume fraction index from 0 to 20 are considered with n=0,
represents the homogenous case of ceramic plate.
Table 4.38: Natural frequencies of simply supported four-parameter FGM plate for first six
modes (h=0.1m)
n
Profiles Mode
0 0.2 0.4 0.8 1 5 10 20
1 3132.852 3125.462 3117.48 3100.12 3090.867 2793.455 1963.022 635.5957
2 7481.315 7465.368 7447.947 7409.564 7388.901 6703.212 4757.477 1565.769
3 7481.315 7465.368 7447.947 7409.564 7388.901 6703.212 4757.477 1565.769
Classic 4 10577.5 10575.27 10574.15 10575.1 10577.09 10339.33 7397.019 2469.472
5 10577.5 10575.27 10574.15 10575.1 10577.09 10755.76 9086.254 3059.584
6 11494.76 11472.3 11447.52 11392.29 11362.3 10755.76 9086.254 3059.584
1 3132.852 3139.264 3144.812 3152.845 3155.066 2735.686 1354.249 234.2482

2 7481.315 7494.977 7506.619 7522.759 7526.66 6537.607 3287.879 578.2702
3 7481.315 7494.977 7506.619 7522.759 7526.66 6537.607 3287.879 578.2702
Symmetric 4 10577.5 10580.63 10584.62 10595.16 10601.7 10049.83 5118.174 913.7849
5 10577.5 10580.63 10584.62 10595.16 10601.7 10853.49 6290.415 1133.513
6 11494.76 11513.77 11529.73 11550.91 11555.25 10853.49 6290.415 1133.513
1 3132.852 3128.547 3123.524 3111.515 3104.584 2788.14 1838.044 510.3354

2 7481.315 7471.966 7460.891 7434.01 7418.333 6684.272 4453.685 1257.335
3 7481.315 7471.966 7460.891 7434.01 7418.333 6684.272 4453.685 1257.335
Asymmetric
4 10577.5 10576.15 10575.84 10578.25 10580.93 10302.35 6922.942 1983.185
5 10577.5 10576.15 10575.84 10578.25 10580.93 10773.04 8502.296 2457.191
6 11494.76 11481.52 11465.62 11426.51 11403.5 10773.04 8502.296 2457.191
Table 4.39: Natural frequencies of simply supported four-parameter FGM plate for first six
modes (h=0.01m)
n
Profiles Mode
0 0.2 0.4 0.8 1 5 10 20
1 324.1771 323.3509 322.4659 320.5594 319.5509 287.9179 200.7168 64.20281
2 810.0952 808.0331 805.8241 801.0643 798.5465 719.5333 501.6757 160.5009
3 810.0952 808.0331 805.8241 801.0643 798.5465 719.5333 501.6757 160.5009
Classic 4 1295.472 1292.178 1288.65 1281.045 1277.022 1150.724 802.4174 256.7668
5 1619.64 1615.526 1611.117 1601.616 1596.589 1438.734 1003.339 321.1017
6 1619.64 1615.526 1611.117 1601.616 1596.589 1438.734 1003.339 321.1017
1 324.1771 324.9004 325.5331 326.4757 326.7571 282.8979 138.2552 23.62688

2 810.0952 811.9001 813.4788 815.8295 816.5307 706.9492 345.567 59.06632
3 810.0952 811.9001 813.4788 815.8295 816.5307 706.9492 345.567 59.06632
Symmetric 4 1295.472 1298.354 1300.875 1304.626 1305.744 1130.537 552.74 94.49552
5 1619.64 1623.24 1626.388 1631.072 1632.467 1413.444 691.1558 118.1739
6 1619.64 1623.24 1626.388 1631.072 1632.467 1413.444 691.1558 118.1739
1 324.1771 323.6979 323.145 321.8382 321.0901 287.5848 187.9573 51.54532

2 810.0952 808.8992 807.5188 804.2559 802.3878 718.6918 469.7836 128.8586
3 810.0952 808.8992 807.5188 804.2559 802.3878 718.6918 469.7836 128.8586
Asymmetric
4 1295.472 1293.561 1291.356 1286.143 1283.157 1149.364 751.4057 206.1462
5 1619.64 1617.253 1614.498 1607.983 1604.253 1437.022 939.5536 257.798
6 1619.64 1617.253 1614.498 1607.983 1604.253 1437.022 939.5536 257.798
From Tables 4.38 and 4.39, it is manifested that the elevation of volume fraction index
from homogeneous to FGM plate increases the frequency of Al/ZrO2 plate. This trend is
186
observed as common phenomenon in all the three profiles (classical, symmetric and
asymmetric). The reason attributed is that the increase in metal content corresponds to low
stiffness thus reducing the frequency as the power law exponent rises.
Table 4.40: Natural frequencies of clamped four-parameter FGM plate for the first six modes
(h=0.1m)
Profiles Mode n
0 0.2 0.4 0.8 1 5 10 20
1 5351.332 5340.513 5328.595 5302.096 5287.73 4800.356 3412.43 1129.664
2 10228.3 10210.51 10190.54 10145.17 10120.19 9231.368 6643.842 2248.666
3 10228.3 10210.51 10190.54 10145.17 10120.19 9231.368 6643.842 2248.666
Classic 4 14361.95 14339.35 14313.63 14254.41 14221.47 13013.22 9442.17 3246.746
5 16951.14 16926.57 16898.25 16832.25 16795.22 15401.39 11234.82 3903.099
6 17111.97 17086.86 17057.97 16990.75 16953.08 15539.9 11324.51 3928.501
1 5351.332 5360.513 5368.242 5378.583 5380.771 4667.175 2355.196 417.5748

2 10228.3 10243.06 10255.12 10269.78 10271.63 8927.615 4593.57 833.7104
3 10228.3 10243.06 10255.12 10269.78 10271.63 8927.615 4593.57 833.7104
Symmetric 4 14361.95 14380.41 14395.17 14411.74 14412.54 12546.16 6535.852 1206.544
5 16951.14 16970.9 16986.37 17002.31 17001.59 14812.36 7780.121 1452.583
6 17111.97 17132.21 17148.1 17164.67 17164.16 14948.57 7840.053 1461.758
1 5351.332 5344.962 5337.332 5318.609 5307.609 4783.676 3192.332 907.0319

2 10228.3 10217.71 10204.7 10172.01 10152.5 9188.311 6212.726 1805.685
3 10228.3 10217.71 10204.7 10172.01 10152.5 9188.311 6212.726 1805.685
Asymmetric
4 14361.95 14348.39 14331.45 14288.21 14262.16 12943.47 8827.05 2607.327
5 16951.14 16936.28 16917.44 16868.71 16839.1 15310.54 10499.89 3134.476
6 17111.97 17096.81 17077.61 17028.05 16997.97 15449.07 10583.46 3154.837
Table 4.41: Natural frequencies of clamped four-parameter FGM plate for first six modes
(h=0.01m)
Profile Mode n
0 0.2 0.4 0.8 1 5 10 20
1 590.5254 589.0234 587.4141 583.9463 582.1116 524.5195 365.718 117.0151
2 1203.612 1200.557 1197.284 1190.227 1186.493 1069.19 745.6393 238.6523
3 1203.612 1200.557 1197.284 1190.227 1186.493 1069.19 745.6393 238.6523
Classic 4 1773.148 1768.657 1763.843 1753.464 1747.97 1575.276 1098.792 351.7923
5 2157.098 2151.64 2145.789 2133.172 2126.494 1916.483 1336.935 428.1054
6 2167.481 2161.996 2156.116 2143.437 2136.726 1925.694 1343.342 430.1467
1 590.5254 591.84 592.9895 594.7007 595.2106 515.3173 251.9106 43.06357

2 1203.612 1206.285 1208.622 1212.097 1213.131 1050.323 513.6197 87.83163
3 1203.612 1206.285 1208.622 1212.097 1213.131 1050.323 513.6197 87.83163
Symmetric 4 1773.148 1777.077 1780.51 1785.613 1787.127 1547.327 756.9039 129.4756
5 2157.098 2161.872 2166.043 2172.241 2174.078 1882.394 920.9689 157.5654
6 2167.481 2172.279 2176.472 2182.701 2184.548 1891.457 925.381 158.3162
1 590.5254 589.6541 588.6485 586.2709 584.9095 523.9001 342.465 93.94575

2 1203.612 1201.84 1199.794 1194.954 1192.183 1067.903 698.2254 191.6026
3 1203.612 1201.84 1199.794 1194.954 1192.183 1067.903 698.2254 191.6026
Asymmetric
4 1773.148 1770.543 1767.533 1760.413 1756.334 1573.345 1028.915 282.4378
5 2157.098 2153.931 2150.273 2141.617 2136.658 1914.115 1251.913 343.7065
6 2167.481 2164.299 2160.622 2151.924 2146.941 1923.318 1257.913 345.3453
187
Table 4.42: Natural frequencies of simply supported-clamped four-parameter FGM plate for
the first six modes (h=0.1m)
Profile Mode n
0 0.2 0.4 0.8 1 5 10 20
1 4156.973 4147.89 4137.986 4116.213 4104.511 3718.128 2628.357 860.5396
2 8792.393 8775.471 8756.735 8714.845 8692.041 7908.192 5654.128 1887.878
3 8847.72 8830.601 8811.659 8769.338 8746.311 7956.141 5685.865 1897.104
Classic 4 12051.01 12048.47 12047.17 12048.19 12050.42 11642.45 8391.26 2844.406
5 12893.85 12871.23 12845.91 12788.57 12757.06 12251.44 10163.68 3476.406
6 14756.33 14753.2 14751.57 14752.63 14755.23 14036.13 10185.23 3482.439
1 4156.973 4164.77 4164.77 4180.733 4183.034 3627.797 1813.751 317.6164

2 8792.393 8806.671 8806.671 8834.216 8837.217 7678.918 3908.736 698.6065
3 8847.72 8862.175 8862.175 8890.135 8893.226 7726.793 3930.362 701.9553
Symmetric 4 12051.01 12054.58 12054.58 12071.05 12078.47 11268.46 5807.666 1054.814
5 12893.85 12912.67 12912.67 12947.22 12949.89 12360.48 7037.288 1290.834
6 14756.33 14760.69 14760.69 14780.67 14789.62 13555.94 7051.744 1293.01
1 4156.973 4151.655 4145.37 4130.151 4121.291 3708.067 2459.947 690.9498

2 8792.393 8782.399 8770.349 8740.596 8723.042 7878.252 5290.149 1515.99
3 8847.72 8837.613 8825.438 8795.399 8777.686 7926.335 5319.885 1523.394
Asymmetric
4 12051.01 12049.47 12049.1 12051.78 12054.79 11589.97 7848.985 2284.268
5 12893.85 12880.4 12863.95 12822.73 12798.19 12270.74 9504.534 2791.885
6 14756.33 14754.43 14753.93 14757.04 14760.59 13966.23 9524.651 2796.722
Table 4.43: Natural frequencies of simply supported-clamped four-parameter FGM plate for
the first six modes (h=0.01m)
Profiles Mode n
0 0.2 0.4 0.8 1 5 10 20
1 444.1443 443.0135 441.802 439.1917 437.8109 394.4827 275.0275 87.98444
2 993.2975 990.7727 988.0675 982.2374 979.1529 882.3082 615.2365 196.8725
3 997.3885 994.8532 992.1366 986.2823 983.1849 885.9395 617.7652 197.6798
Classic 4 1522.186 1518.323 1514.183 1505.26 1500.538 1352.21 943.0533 301.8472
5 1878.978 1874.214 1869.109 1858.101 1852.276 1669.238 1164.262 372.7016
6 1881.43 1876.66 1871.547 1860.525 1854.692 1671.413 1165.775 373.1838
1 444.1443 445.1342 446.000 447.2893 447.6739 387.5844 189.4415 32.37916

2 993.2975 995.5071 997.4393 1000.315 1001.171 866.8104 423.7926 72.45336
3 997.3885 999.6074 1001.548 1004.435 1005.295 870.3809 425.5342 72.75036
Symmetric 4 1522.186 1525.566 1528.52 1532.915 1534.222 1328.358 649.6201 111.0897
5 1878.978 1883.147 1886.79 1892.207 1893.817 1639.73 802.0132 137.1686
6 1881.43 1885.604 1889.252 1894.677 1896.289 1641.87 803.0553 137.346
1 444.1443 443.4884 442.7314 440.9421 439.9176 394.0218 257.5425 70.63843

2 993.2975 991.833 990.1423 986.145 983.856 881.2617 576.1202 158.0596
3 997.3885 995.9178 994.2201 990.2061 987.9076 884.8894 578.4883 158.7077
Asymmetric
4 1522.186 1519.945 1517.357 1511.238 1507.733 1350.583 883.0914 242.3392
5 1878.978 1876.215 1873.023 1865.474 1861.15 1667.216 1090.233 299.2252
6 1881.43 1878.663 1875.467 1867.907 1863.578 1669.389 1091.65 299.6123
Further for all the three profiles considered, the value of n= 0, produces the same
frequency, due to the isotropic property of plate. In Table 4.32, the symmetric profile
exhibits high frequency value followed by asymmetric and classical profiles, when the
power law exponent ranges from 0 to 1. Beyond the linear range (n>1), the classical FGM
188
plates produces higher frequency, thus ensuring the high stiffness of the plate under
consideration. The low values of frequencies are recorded for thin plate compared to
moderately thick plates as expected for different values of volume fraction index. The
observations regarding the influence of chosen profile on frequency extracted from Table
4.38 are analogous to Table 4.39. The symmetric profile of FGM plate ensures the ceramic
segment at top and bottom of the plate having high stiffness and shows better performance
compared to other two cases of FGM profiles.
Free vibration results of Al/ZrO2 thin and moderately thick plate with clamped
boundary is furnished in Tables 4.40 and 4.41. Because of the high bending stiffness of the
clamped boundary, the higher values of frequency values are reported for both the cases.
The observations regarding the profile type on natural frequencies in Tables 4.38 and 4.39
hold true for Table 4.40 and 4.41 also, except the frequency values are higher for the latter
case. In Tables 4.42 and 4.43, simply supported-clamped FGM plates are considered to
generate the frequency values. Intermediate values of frequency are recorded, since two of
the edges correspond to simply-supported boundary, thereby reducing the total stiffness of
the plate. Once again, symmetric profile is observed to be a better choice compared to
classical and asymmetric profiles, due to higher stiffness values at top and bottom of the
plate.
4.5.2.2 Influence of material gradient parameters on free vibration

analysis of four-parameter Al/ZrO2 plate
Influences of various parameters that define the material gradient profile in four-parameter
power law are depicted in Figures 4.27, 4.28 and 4.29. In Figure 4.26, free vibration
results of Al/ZrO2 FGM plate are shown, in which the parameters b1 and c1 are kept
constant and the parameter a1 is varied from 0 to 1.2. In all the cases, a ceramic line is
established which represents high stiffness region of the plate. A fast descending behavior
of frequency is observed in case of FGM plates compared to isotropic plates (n=0).
Increase in the value of volume fraction index tends to reduce the stiffness of the plate. In
some cases, the natural frequency of FGM plate exceeds frequency of ceramic plate. For
example, in frequency mode 5 and 6, lower value of the parameter a1 (0.2 to  0.8) exceeds
the maximum frequency of the plate. This is due to the choice of parameters b1 and c1 to
decide the frequency value. In particular, the types of vibration mode that ensures this type
of monotonous decrease of frequency are torsional, bending and axi-symmetric mode
189
shapes. In Figure 4.28, the parameter b1 is varied from 0 to 1.2 while other two parameters
(a1 and c1) are kept constant.
3500
ceramic line
8000
3000 ceramic line
a=0
a=0.2 a=0
2500
a=0.4 6000 a=0.2
Frequency (
a=0.6 a=0.4
Frequency (
2000 a=0.8 a=0.6

a=1.0 a=0.8
a=1.2 4000
1500 a=1.0
a=1.2
1000
2000
500
0 0
0 20 40 60 80 100 0 20 40 60 80 100

(a) Mode 1 (b) Mode 2 and 3
12000 12000
10000 10000 a=0

a=0
a=0.2 a=0.2
a=0.4 8000 a=0.4
Frequency (
8000 a=0.6
a=0.6
Frequency (
a=0.8 a=0.8
6000 a=1.0
6000 a=1.0
a=1.2 a=1.2
4000
4000
2000
2000
0
0 20 40 60 80 100 0 20 40 60 80 100
(c) Mode 4 (d) Mode 5
12000 ceramic line
a=0
10000
a=0.2
a=0.4
Frequency (
8000 a=0.6
a=0.8
a=1.0
6000
a=1.2
4000
2000
0
0 20 40 60 80 100
Volume fraction index (n)

(e) Mode 6
Figure 4.27 First six natural frequencies of four-parameter Al/ZrO2 FGM plate with varying
b1 and c1 values (0<a1<1.2)
190
3500
8000
3000
7000
2500 6000
b=0
b=0.2
Frequency (
b=0
5000 b=0.4
2000 b=0.2
Frequency (
b=0.6
b=0.4 4000 b=0.8
1500 b=0.6 b=1.0
b=0.8 3000 b=1.2
b=1.0
1000
b=1.2 2000
500 1000
0
0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
12000
(a) Mode 1 12000 (b) Mode 2 and 3
10000 10000
b=0
8000 b=0.2 8000
b=0
Frequency (
Frequency (
b=0.4 b=0.2
b=0.6 b=0.4
6000 6000 b=0.6
b=0.8
b=1.0 b=0.8
4000 b=1.2 4000
b=1.0
b=1.2
2000 2000
0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60

12000
10000
b=0
Frequency (
8000 b=0.2
b=0.4
6000
b=0.6
b=0.8
b=1.0
4000
b=1.2
2000
0 10 20 30 40 50 60
(e) Mode 6
a1 and c1 values (0<b1<1.2)
A convex type of descending behavior is discerned in all the type of frequency modes. For
homogeneous case of plate, all the cases merge at same frequency value. Beyond certain
value of volume fraction index (say n  25), the frequency of plate considering different
191
values of the parameter b1 establishes stable path. Hence it can be inferred that change in
value of the parameter b1 has no significant effect beyond certain value of n.
3500
ceramic line 8000 ceramic line
3000
7000
2500
6000
Frequency (
c=1 c=1
c=3
Frequency (
2000 5000 c=3
c=5 c=5
c=7 4000 c=7
1500
c=9 c=9
c=11 3000 c=11
1000
2000
500
1000
0
0
0 20 40 60 80 100 0 20 40 60 80 100
12000 (a) Mode 1 (b) Mode 2 and 3
12000
10000
10000
8000 c=1 c=1

8000
Frequency (
c=3 c=3
Frequency (
c=5 c=5
6000
c=7 6000 c=7
c=9 c=9
4000 c=11 c=11
4000
2000 2000
0 0
0 20 40 60 80 100 0 20 40 60 80 100
12000 ceramic line
10000
c=1
c=3
Frequency (
8000
c=5
c=7
6000 c=9
c=11
4000
2000
0
0 20 40 60 80 100
(e) Mode 6
a1 and b1 values (1<c1<11)
192
The first six mode shapes of FGM plate for several ranges of the power law
parameter c1 is exhibited in Figure 4.29. For c1=1, the ceramic line corresponds to high
stiffness of the plate. For higher values of c1 (c1=3, 7, 9 and 11), a steep decreasing
tendency of frequency values are noticed. Exceeding the volume fraction index beyond 40,
shows stable point for all the cases of c1 considered, except for c1=1. Further in mode 5, for
c1 = 3, the free vibration frequency value exceeds the frequency of ceramic plate due to the
choice of the other parameters a1 and b1. This behavior depends on the type of vibration
mode and value of the parameter c1.
4.5.3 Buckling of four-parameter FGM plate subjected to mechanical and

thermal loading
Under this section, buckling results are presented for different in-plane loading conditions.
To perform the study, different kinds of FGM profiles are incorporated. In addition,
parametric results are presented for different (h/a) ratio and aspect ratio (a/b) values. The
following non-dimensional buckling load parameter is used to represent the critical
buckling load.
N cr b 2 Ec h3
Buckling load: N cr  (where D  )
 2 Dc 12(1  2 )
c
4.5.3.1 Buckling of four-parameter Al/Al2O3 plate subjected to

mechanical loading
In this section, buckling results of Al/Al2O3 are presented by considering various geometric
parameters subjected to uni-axial compression, bi-axial compression and bi-axial
compression and tension as shown in Figure 4.30.
Figure 4.31 shows the variation of critical buckling load parameter with volume
fraction index (n) subjected to uni-axial compression, bi-axial compression and bi-axial
compression and tension, for square Al/Al2O3 thick plate (h=0.1m), under simply
supported boundary. All the results are generated for linear variation of volume fraction
index (n=1.0) of the power law function. Value of volume fraction index (n) was chosen in
the range of 0 to 10.0 to represent homogenous and graded properties of the plate. In case
of FGM2, with increase in volume fraction index up to n=2.0 critical buckling load
decreases, beyond which it starts to increase and attains maximum value. On the contrary,
FGM1 exhibit monotonic rising tendency of critical load with respect to various volume
fraction indexes. Since n=0.0 indicates the homogenous plate case, FGM with different
193
material profiles merge at the same point. Further, more load is necessary to buckle the
symmetric plates (FGM2) compared to FGM1 plates. As expected, critical load
corresponding to uni-axial compression is greater than bi-axial compression and less than
that of bi-axial compression and tension.
Py
Px Px
b
Px Px
b
a
a
Py
(a) Uni-axial (b)Bi-axial compression
z compression
Py
x
y
Px Px
b
Py
(c) Bi-axial compression

and tension
Figure 4.30 FGM plate subjected to various in-plane loading conditions
Next, the influence of thickness of the plate on critical buckling load was studied in Figure
4.32. Thickness range was chosen appropriately to examine the plate under thin and thick
category. As visualized, when thickness increases decreasing tendency is observed to
decrease for both the types of profiles. Influence of aspect ratio on critical buckling load is
shown for two types of profiles in Figure 4.33.
194
8.0
4.0
Non-dimensional buckling load (Ncr)
7.5

3.6
7.0
6.5
3.2 FGM1
FGM1 FGM2
6.0
FGM2
5.5 2.8
5.0
2.4
4.5
4.0 2.0
3.5
1.6
0 2 4 6 8 10 0 2 4 6 8 10
(a) Uni-axial compresssion (b) Bi-axial compresssion
15
14
13
12 FGM1
FGM2
11
10
6
0 2 4 6 8 10
(c) Bi-axial compresssion and tension
Figure 4.31 Variation of critical buckling load (Ncr) of four-parameter Al/Al2O3 plate
for different volume fraction index (n) under different conditions of in-plane edge loads
4.4
4.2
4.0
3.8
3.6
3.4
FGM1
3.2 FGM2
3.0
0.00 0.05 0.10 0.15 0.20

Thickness-side ratio (h/a)
Figure 4.32 Influence of thickness-side (h/a) ratio on critical buckling load (Ncr) of four-
parameter Al/Al2O3 plate subjected to bi-axial compression
195
5.6

5.2
4.8
FGM1
FGM2
4.4
4.0
3.6
0 1 2 3 4 5
Aspect ratio (a/b)
Figure 4.33 Influence of aspect ratio (a/b) on critical buckling load (Ncr) of four-parameter
Al/Al2O3 plate subjected to bi-axial compression
Critical buckling variation is not smooth due to the change of buckling value and this trend
is observed up to a/b=2.0, beyond that asymptotic response is seen. Hence, it can be
inferred that a/b value between 1.0 and 2.0 reflects marked response on critical buckling
load under linear variation of volume fraction index. Hence it can be concluded that by
suitable assumption of different parameters in power law distribution, various design
requirements imposed on FGM structure can be satisfied.
4.5.3.2 Buckling load for four-parameter Si3N4/SUS3O4 plate subjected to
thermal loading
Having studied and discussed the response of different FGM models under various
mechanical edge loads in section 4.5.3.1, effect of temperature field on buckling load has
been studied in this section. For this case homogenization is applied only for Young’s
modulus (E) and co-efficient of thermal expansion (α). These parameters are considered to
be dependent on position (z) and temperature (T). Thermal conductivity (k) is assumed to
be position (z) dependent and temperature (T) independent. Value of Poisson’s co-efficient
is assumed to be constant, since it hardly depends on any temperature effect encountered in
the problem. Critical buckling temperature (∆Tcr) of FGM models under different
boundary conditions is furnished in Table 4.44 for square Si3N4/SUS3O4 plate with n=1.
Thickness of the plate is assumed to be, h=0.1m. Symmetric profile (FGM2) reports
maximum critical buckling temperature (∆Tcr) and this trend is not influenced by the type
of boundary constraints imposed on the edges. For clamped conditions on all four edges
196
the maximum buckling temperature (∆Tcr) is observed due to high bending rigidity offered
by the plate.
Table 4.44: Critical buckling temperature (∆Tcr) of four-parameter Si3N4/SUS304 plate under
Critical buckling temperature (∆Tcr)
Models
SSSS CCCC SCSC CFCF
FGM1 1603.12 3811.27 2466.84 679.59
FGM2 1799.41 4282.17 2768.79 764.98
To study the non-linear distribution of temperature field (T) across the thickness,
temperature difference (∆T) between ceramic and metal is chosen from 200°C to 1200°C.
Simply supported boundary with a/b=1.0 and linear variation of volume fraction index is
chosen for the present example (Table 4.45).
As the temperature rises, the critical buckling temperature (∆Tcr) falls considerably,
due to the low value of Young’s modulus of the plate at high temperature. Among various
FGM modes, FGM2 gives the maximum critical buckling temperature (∆Tcr) in this group,
as observed in former cases also.
Table 4.45: Critical buckling temperature (∆Tcr) of four-parameter Si3N4/SUS3O4 plate under
non-linear temperature field (°C)
Critical buckling temperature (∆Tcr)
Model
200 400 600 800 1000
FGM1 18.025 9.012 6.008 4.506 3.605
FGM2 20.778 10.389 6.926 5.194 4.155
4.6 STATIC ANALYSIS OF FGM SANDWICH SKEW PLATES AND

SHELLS
4.6.1 Static analysis of FGM sandwich skew plates subjected to
mechanical loading
Under this section, wide varieties of functionally graded skew sandwich plates are
considered to perform the numerical analysis. Each kind of sandwich plate is identified by
thickness parameter which is designated from the bottom to the top of the plate. Due to
lack of data for functionally graded skew sandwich plates (FGSSP) in the literature, the
present FE formulation is validated and assessed for functionally graded sandwich plates
(FGSP) without skew. In this connection, new results are generated and important findings
197
are discussed with respect to bending analysis of functionally graded skew sandwich plates
constituted of Al/ZrO2 combination by considering different values of skew angles ().
Parametric studies are carried out by assuming various thickness ratio (a/h), aspect ratio
(a/b), boundary conditions and power law exponent (n) of FGSSP having different
combinations of bottom-core-top thickness with the choice of core/isotropic material in the
core layer.
For static analysis, unless mentioned otherwise, the results are presented in the
following non-dimensional form:
For FGSSP with homogeneous core (Type A):
 a b  10hE  a b  10h  a b  10h

2 2
w  ,   2 0 w  where E0  1GPa  ,  xx  ,   2  xx ,  yy  ,   2  yy
 2 2  a qz  2 2  a qz  2 2  a qz
h  b h  a h
and  xy  0, 0    xy ,  xz  0,    xz ,  yz  0,    yz .
aqz  2  aqz  2  aqz
For FGSSP with FGM core (Type B):
 a b  10h Ec a b h a b h
3
w ,   4 w ,  xx  ,    xx ,  yy  ,    yy
 2 2  a qz  2 2  aqz  2 2  aqz
h  b h  a h
and  xy  0, 0    xy ,  xz  0,    xz ,  yz  0,    yz .
aqz  2  aqz  2  aqz
4.6.1.1 Convergence and validation study for FGM sandwich plate

subjected to mechanical loading
Example 1: As a first validation example, Al/ZrO2 FGSP is considered with FGM layer
at top and bottom face sheets and homogeneous ceramic material as core (Type A). Type
A plate with simply supported boundary condition having thickness ratio a/h=10 is taken
to generate a table. Deflection results are compared for FGSP having five kinds of
thickness schemes, viz. 2-1-2, 2-1-1, 1-1-1, 2-2-1 and 1-2-1 and for several values of
power law exponent (n=0, 0.2, 0.5, 1.0, 2.0, 5.0 and 10). Values of non-dimensional
displacement (w) of Type A plate are furnished in Table 4.46 and these results are
validated with the results reported by Zenkour (2005a) based on analytical solution and
Neves et al. (2012 a) based on radial basis meshless collocation technique. Percentage of
error of deflection results between present study and Neves et al. (2012 a) are also shown
in Table 4.46. The results reported by Zenkour (2005a), based on various shear
198
Table 4.46: Non-dimensional displacements for Type A Al/ZrO2 FGSP for different thickness
schemes and volume fraction indexes (n)
Thickness scheme
n References
2-1-2 2-1-1 1-1-1 2-2-1 1-2-1
0 Zenkour (2005a) SSDT 0.19605 - 0.19605 0.19605 0.19605
Zenkour (2005a)TSDT 0.19606 - 0.19606 0.19606 0.19606
Zenkour (2005a) FSDT 0.19607 - 0.19607 0.19607 0.19607
Zenkour (2005a)CLPT 0.18560 - 0.18560 0.18560 0.18560
Neves et al. (2012a) 0.1949 0.1949 0.1949 0.1949 0.1949
Present (6x6)a 0.1980 0.1980 0.1980 0.1980 0.1980
% of error 1.59 1.59 1.59 1.59 1.59
0.2 Neves et al. (2012a) 0.2297 0.2275 0.2261 0.2235 0.2209
Present 0.2269 0.2027 0.2245 0.2050 0.2232
% of error 1.22 0.11 0.71 8.27 1.04
0.5 Neves et al. (2012a) 0.2650 0.2597 0.2566 0.2503 0.2444
Present 0.2644 0.2524 0.2573 0.2475 0.2498
% of error 0.23 2.81 0.27 1.11 2.20
1.0 Zenkour (2005a) SSDT 0.30624 - 0.29194 0.28082 0.27093
Zenkour (2005a) TSDT 0.30632 - 0.29199 0.28085 0.27094
Zenkour (2005a) FSDT 0.30750 - 0.29301 0.28168 0.27167
Zenkour (2005a) CLPT 0.29417 - 0.28026 0.26920 0.25958
Neves et al. (2012a) 0.3070 0.2975 0.2929 0.2820 0.2722
Present 0.3093 0.3028 0.2965 0.2887 0.2812
% of error 0.75 1.78 1.23 2.38 3.31
2.0 Zenkour (2005a) SSDT 0.35218 - 0.33280 0.31611 0.30260
Zenkour (2005a) TSDT 0.35231 - 0.33289 0.31617 0.30263
Zenkour (2005a) FSDT 0.35408 - 0.33441 0.31738 0.30370
Zenkour (2005a) CLPT 0.33942 - 0.32067 0.30405 0.29095
Neves et al. (2012a) 0.3519 0.3376 0.3329 0.3164 0.3032
Present 0.3560 0.3103 0.3386 0.2999 0.3157
% of error 1.16 8.08 1.71 5.22 4.12
5 Zenkour (2005a) SSDT 0.39160 - 0.37128 0.34950 0.33474
Zenkour (2005a) TSDT 0.39183 - 0.37145 0.34960 0.33480
Zenkour (2005a) FSDT 0.39418 - 0.37356 0.35123 0.33631
Zenkour (2005a) CLPT 0.37789 - 0.35865 0.33693 0.32283
Neves et al. (2012a) 0.3905 0.3772 0.3705 0.3490 0.3347
Present 0.3967 0.3451 0.3788 0.3332 0.3514
% of error 1.59 8.51 2.24 4.53 4.99
10 Zenkour (2005a) SSDT 0.40376 - 0.38490 0.34916 0.34119
Zenkour (2005a) TSDT 0.40407 - 0.38551 0.36215 0.34824
Zenkour (2005a) FSDT 0.40657 - 0.38787 0.36395 0.34996
Zenkour (2005a) CLPT 0.38941 - 0.37236 0.34915 0.33612
Neves et al. (2012a) 0.4026 0.3835 0.3843 0.3612 0.3480
Present 0.4079 0.3562 0.3935 0.3662 0.3670
% of error 1.32 7.12 2.39 1.38 5.45
SSDT: sinusoidal shear deformation plate theory; TSDT: third order shear deformation plate
theory; FSDT: first order shear deformation plate theory; CLPT: classical layer plate theory
a
mesh division
.
deformation theories, viz. CPT, FSDT, TSDT and SSDT are compared with the present
results. However, the variation of transverse displacement is not considered in the
199
displacement field of Zenkour (2005a) and assumed to be constant throughout the
thickness.
In addition, results based on meshless technique (Neves et al. (2012 a)
incorporating thickness stretching effect in the displacement field is also considered. It is
concluded that for majority of the cases, the present results overestimate the displacement
component compared with the reference results especially in comparison with the
analytical solutions (Zenkour 2005a). Highest percentage of deviation with respect to the
analytical solution is found to be about 8.85% for the n = 5.0 and 1-2-1 thickness scheme.
It is an expected trend, because the analytical model has not considered the variation of
transverse displacement and hence results in a stiffer model. In addition, there also some
small variation in the present results compared to meshless technique. This may be owing
to different solution methodologies adopted in the studies.
Table 4.47: Validation table for Type B Al/Al2O3 FGM sandwich plate having different
thickness schemes and various volume fraction index (n)
n References a/h
4 10 100
0 Neves et al. (2013b) 0.4447 0.3711 0.3568
Present (6x6)a 0.4710 0.3807 0.3636
% of error 5.91 2.59 1.90
0.5 Neves et al. (2013b) 0.6168 0.5238 0.5058
Present 0.6927 0.5758 0.5537
% of error 12.30 9.92 9.47
1.0 Brischetto (2009) 0.7628 0.6324 0.6072
Carrera et al. (2011) 0.7628 0.6324 0.6072
Neves et al. (2012c) 0.7418 0.6305 0.6092
Neves et al. (2013b) 0.7417 0.6305 0.6092
Present 0.7803 0.6368 0.6096
% of error 5.20 0.99 0.06
4.0 Brischetto (2009) 1.0934 0.8321 0.7797
Carrera et al. (2011) 1.0930 0.8307 0.7797
Neves et al. (2012c) 1.0391 0.8202 0.7784
Neves et al. (2013b) 1.0371 0.8199 0.7784
Present 1.0384 0.8444 0.7901
% of error 0.13 2.99 1.51
10 Brischetto (2009) 1.2232 0.8753 0.8077
Carrera et al. (2011) 1.2172 0.8740 0.8077
Neves et al. (2012c) 1.1780 0.8650 0.8050
Neves et al. (2013b) 1.1752 0.8645 0.8050
Present 1.1071 0.8889 0.8690
% of error 5.79 2.82 7.95
Example 2: In the second example, to validate the results for functionally graded
sandwich plate having FGM material as core segment (Type B), Aluminium and Alumina
constituents are considered. For comparison, sandwich plate with 1-8-1 thickness scheme
having simply supported boundary conditions at all the edges is taken. The non-
dimensional displacement parameters (w) for Type B FGSP for several values of volume
200
fraction index n= 0, 0.5, 1.0, 4.0 and 10.0 are shown in Table 4.47. The deflection results
obtained are compared with Brischetto (2009), Carrera et al. (2011), Neves et al., (2012 c)
and Neves et al. (2013b), which are based on radial basis collocation technique (meshless
technique). From Table 4.47, it is observed that the present results are reasonably close to
the referred results.
4.6.1.2 Influence of skew angle () and volume fraction index (n) on
displacement parameter for Type A FGSSP
Table 4.48: Non-dimensional central deflection of Type A FGSSP for various skew
angle () and volume fraction index (n) (a/h=4)
Thickness scheme
n 
2-1-2 2-1-1 1-1-1 2-2-1 1-2-1
0 15 0.0353 0.0327 0.0355 0.0339 0.0362
30 0.0197 0.0184 0.0198 0.0190 0.0202
45 0.0047 0.0044 0.0047 0.0045 0.0048
60 -0.0018 -0.0017 0.0018 -0.0017 -0.0018
0.2 15 0.0397 0.0365 0.0394 0.0373 0.0394

30 0.0220 0.0204 0.0219 0.0209 0.0219
45 0.0052 0.0048 0.0052 0.0049 0.0051
60 -0.0019 -0.0018 0.0019 -0.0019 -0.0019
0.5 15 0.0453 0.0439 0.0442 0.0437 0.0434

30 0.0250 0.0243 0.0244 0.0242 0.0241
45 0.0058 0.0057 0.0057 0.0056 0.0057
60 -0.0021 -0.0021 0.0021 -0.0021 -0.0020
1.0 15 0.0520 0.0516 0.0501 0.0499 0.0482

30 0.0285 0.0283 0.0275 0.0269 0.0265
45 0.0066 0.0066 0.0064 0.0064 0.0062
60 -0.0023 -0.0024 0.0023 -0.0023 -0.0022
2.0 15 0.0592 0.0535 0.0564 0.0520 0.0532

30 0.0323 0.0294 0.0308 0.0287 0.0292
45 0.0074 0.0068 0.0071 0.0066 0.0067
60 -0.0026 -0.0025 0.0025 -0.0024 -0.0024
5.0 15 0.0658 0.0595 0.0626 0.0576 0.0585

30 0.0359 0.0327 0.0341 0.3168 0.0320
45 0.0083 0.0075 0.0078 0.0073 0.0074
60 -0.0029 -0.0028 0.0027 -0.0026 -0.0026
10 15 0.0682 0.0619 0.0651 0.0600 0.0610

30 0.0373 0.0341 0.0355 0.0330 0.0333
45 0.0086 0.0079 0.0081 0.0076 0.0076
60 -0.0030 -0.0029 0.0028 -0.0027 -0.0027
Under this section, it is aimed to generate new results for FGSSP for different values of
skew angles (). For all the numerical examples performed, Al/ZrO2 combination is
201
considered. In addition, parametric study is shown by varying material and geometric
parameters. Non-dimensional deflections for Type A FGSSP having thickness ratio a/h=4,
10 and 100 are presented in Table 4.48, 4.49 and 4.50, respectively, to sketch the influence
of the skew angle (), volume fraction index (n) and thickness ratio (a/h) on bending
response of sandwich skew plates. Different values of volume fraction index (n=0, 0.2, 0.5,
1.0, 2.0, 5.0 and 10) are also chosen. For each case a set of skew angles ranging from 15
to 60 are incorporated.
Table 4.49: Non-dimensional central deflection of Type A FGSSP for various skew
angle () and volume fraction index (n) (a/h=10)
Thickness scheme
n 
2-1-2 2-1-1 1-1-1 2-2-1 1-2-1
0 15 0.1614 0.1461 0.1620 0.1497 0.1647
30 0.0808 0.0737 0.0811 0.0756 0.0827
45 0.0169 0.0154 0.0170 0.01596 0.0174
60 -0.0049 -0.0046 -0.0049 -0.0047 -0.0051
0.2 15 0.1843 0.1656 0.1824 0.1675 0.1819

30 0.0915 0.0829 0.0907 0.0840 0.0907
45 0.0190 0.0172 0.0189 0.0176 0.01896
60 -0.0055 -0.0050 -0.0054 -0.0052 -0.0055
0.5 15 0.2139 0.2057 0.2083 0.2011 0.2030

30 0.1051 0.1009 0.1026 0.0995 0.1006
45 0.0216 0.0206 0.0212 0.0206 0.0209
60 -0.0062 -0.0060 -0.0061 -0.0059 -0.0060
1.0 15 0.2490 0.2441 0.2391 0.2333 0.2277

30 0.1213 0.1189 0.1167 0.1144 0.1120
45 0.0247 0.0240 0.0238 0.0234 0.0230
60 -0.0071 -0.0069 -0.0068 -0.0067 -0.0066
2.0 15 0.2857 0.2505 0.2719 0.2423 0.2547

30 0.1382 0.1223 0.1317 0.1187 0.1244
45 0.0279 0.0248 0.0267 0.0242 0.0253
60 -0.0079 -0.0071 -0.0075 -0.0070 -0.0072
5.0 15 0.3181 0.2784 0.3035 0.2686 0.2827

30 0.1536 0.1358 0.1462 0.1310 0.1371
45 0.0311 0.0275 0.0295 0.0267 0.0277
60 -0.0088 -0.0079 -0.0083 -0.0077 -0.0079
10 15 0.3283 0.2878 0.3153 0.2792 0.2950

30 0.1592 0.1408 0.1520 0.1363 0.1429
45 0.0323 0.0286 0.0307 0.0278 0.0288
60 -0.0092 -0.0083 -0.0087 -0.0080 -0.0082
As the value of skew angle diverges from 15 to 60, a fall-off tendency in the
displacement component is observed for all the thickness schemes. The degree of
hardening increases for higher skew angles, thus results in lower value of transverse
202
displacement component. More interestingly, it can be visualized that Type A FGSSP
records negative displacement component at the acute corners i.e, = 60, which is quite
opposite behavior compared to other cases of skew angles (15, 30 and 45).
Table 4.50: Non-dimensional central deflection of Type A FGSSP for various skew angle ()
and volume fraction index (n) (a/h=100)
Thickness schemes
n 
2-1-2 2-1-1 1-1-1 2-2-1 1-2-1
0 15 5.5300 5.4138 5.5524 5.3039 5.6595
30 1.6326 1.6058 1.6388 1.5735 1.6684
45 0.2671 0.2636 0.2682 0.2576 0.2732
60 -0.0728 -0.0718 -0.0731 -0.0704 -0.0744
0.2 15 6.2179 5.9867 6.1599 5.8085 6.1693

30 1.8290 1.7696 1.8117 1.7172 1.8128
45 0.2997 0.2905 0.2969 0.2814 0.2974
60 -0.0815 -0.0790 -0.0807 -0.0767 -0.0808
0.5 15 7.0853 7.0678 6.9127 6.7245 6.7871

30 2.0762 2.0759 2.0255 1.9762 1.9874
45 0.3409 0.3413 0.3326 0.3244 0.3266
60 -0.0924 -0.0925 -0.0901 -0.0881 -0.0885
1.0 15 8.1126 8.1455 7.7964 7.5999 7.5030

30 2.3693 2.3820 2.2764 2.2238 2.1897
45 0.3898 0.3924 0.3746 0.3658 0.3606
60 -0.1053 -0.1060 -0.1012 -0.0990 -0.0974
2.0 15 9.2046 8.4773 8.7491 7.9089 8.2853

30 2.6826 2.4833 2.5479 2.3153 2.4108
45 0.4419 0.4091 0.4199 0.3809 0.3977
60 -0.1191 -0.1107 -0.1131 -0.1031 -0.1071
5.0 15 10.2484 9.4152 9.7010 8.6809 9.1072

30 2.9864 2.7568 2.8212 2.5376 2.6440
45 0.4920 0.4544 0.4654 0.4179 0.4370
60 -0.1325 -0.1229 -0.1252 -0.1131 -0.1174
10 15 10.6514 9.7955 10.0973 9.0131 9.4854

30 3.1075 2.8713 2.9370 2.6349 2.7520
45 0.5116 0.4732 0.4845 0.4340 0.4550
60 -0.1379 -0.1281 -0.1303 -0.1174 -0.1222
In addition, the effect of n over different types of thickness schemes also can be
studied from these tables. It can be ascertained that, ceramic plate (n=0) exhibit lower
displacement component while graded plates shows larger values of transverse
displacements and it depends on value of volume fraction index. The reason for the above
trend is the high stiffness offered by the ceramic constituent compared to metal constituent.
FGM plate with intermediate properties experience displacement values between those of
Aluminium and Zirconia. For particular value of skew angle () and volume fraction
203
index (n), if equal or higher value of thickness combination is selected for top and bottom
face sheets with respect to core thickness, a considerable reduction in displacement
component is observed. In that sense, FGSSP designated with thickness schemes 2-2-1 and
2-1-1 turns out to be best option over other choices.
4.6.1.3 Influence of skew angle () and volume fraction index (n) on
displacement parameter for Type B FGSSP
Table 4.51 furnishes the values of transverse displacement for type B FGSSP by
employing different skew angles () and volume fraction index (n). Thickness scheme
represented by 1-8-1 notation is considered with thickness ratio a/h= 4, 10 and 100. When
the value of volume fraction index (n) approaches the value from 0 to 10, the deflection
parameter rises due to the less stiffness of the metal segment compared to ceramic part.
The dominance of skew angle () over the displacement component is similar to the one
discussed in case of Type A FGSSP. In general, FGSSP having skew angle value beyond
45 ( i.e., the case of skew plate with acute corner) has the tendency to exhibit negative
deflection parameter for all the values of thickness ratio (a/h) and volume fraction index
(n) for both Type A and Type B FGSSP.
Table 4.51: Non-dimensional transverse displacement of Type B FGSSP for different skew
angle () and volume fraction index (n)
n
a/h 
0 0.2 0.5 1.0 2.0 5.0 10.0
4 15 0.4428 0.4579 0.4816 0.5235 0.6110 0.8029 0.9131

30 0.2469 0.2562 0.2707 0.2964 0.3500 0.4664 0.5306
45 0.0581 0.0605 0.0643 0.0711 0.0851 0.1151 0.1311
60 -0.0213 -0.0225 -0.0245 -0.0279 -0.0351 -0.0503 -0.0572
10 15 0.3301 0.3369 0.3470 0.3637 0.3955 0.4675 0.5381

30 0.1653 0.1696 0.1761 0.1866 0.2062 0.2483 0.2855
45 0.0347 0.0358 0.0374 0.0401 0.0452 0.0559 0.0644
60 -0.0101 -0.0104 -0.0111 -0.0120 -0.0139 -0.0180 -0.0207
100 15 0.1119 0.1150 0.1195 0.1265 0.1382 0.1611 0.1868

30 0.0326 0.0336 0.0350 0.0372 0.0409 0.0482 0.0563
45 0.0054 0.0055 0.0058 0.0061 0.0067 0.0079 0.0092
60 -0.0014 -0.0015 -0.0016 -0.0017 -0.0018 -0.0022 -0.0026
204
4.6.1.4 Influence of boundary conditions on displacement parameter for
Type A and Type B FGSSP
To investigate the influence of aspect ratio (a/b) on the bending response of FGSSP, Type
A and Type B FGSSP are considered based on the criteria that the value of a is fixed at
100 cm and b is varied to investigate the influence of aspect ratio (a/b). Central
displacement (w) for Type A and Type B FGSSP by considering several values of aspect
ratio (a/b) are presented in Table 4.52. To execute the numerical part, various values of
power law exponent n=0, 0.2, 0.5,
1.0, 2.0, 5.0 and 10.0 are considered. Thickness schemes 1-2-1 and 1-8-1 are considered
for type A and type B plate, respectively. From Table 4.52, it is observed that for aspect
ratio (a/b) from 0.5 to 1.5, positive displacement occurs, while FGSSP with aspect ratio
a/b=2.0 and 3.0 records negative deflection parameter for both type A and type B FGSSP.
Further, the influence of higher skew angle is to decrease the deflection parameter for all
the values of aspect ratio (a/b).
Table 4.52: Influence of aspect ratio (a/b) on non-dimensional deflection (w) of Type A and
Type B FGSSP for various skew angle () and volume fraction index (n)
n
a/b  Type of plate
0 0.2 0.5 1.0 2.0 5.0 10.0
0.5 15 Type A 0.0114 0.0116 0.0121 0.0128 0.0143 0.0177 0.0203
Type B 0.0218 0.0240 0.0266 0.0298 0.0332 0.0367 0.0383
30 Type A 0.0040 0.0042 0.0044 0.0047 0.0053 0.0067 0.0077
Type B 0.0077 0.0085 0.0093 0.0104 0.0115 0.0126 0.0131
1.0 15 Type A 0.0868 0.0886 0.0913 0.0957 0.1040 0.1230 0.1415
Type B 0.1647 0.1818 0.2029 0.2277 0.2547 0.2826 0.2950
30 Type A 0.0435 0.0446 0.0463 0.0490 0.0542 0.0653 0.0751
Type B 0.0826 0.0907 0.1005 0.1119 0.1243 0.1371 0.1428
1.5 15 Type A 0.0876 0.0893 0.0918 0.0958 0.1035 0.1212 0.1401
Type B 0.1670 0.1846 0.2063 0.2318 0.2595 0.2882 0.3008
30 Type A 0.0170 0.0174 0.0180 0.0190 0.0208 0.0250 0.0291
Type B 0.0331 0.0364 0.0403 0.0449 0.0498 0.0548 0.0570
2.0 15 Type A -0.0120 -0.0123 -0.0126 -0.0132 -0.0144 -0.0167 -0.0189
Type B -0.0219 -0.0242 -0.0270 -0.0304 -0.0342 -0.0381 -0.0398
30 Type A -0.0502 -0.0514 -0.0530 -0.0557 -0.0608 -0.0720 -0.0829
Type B -0.0953 -0.1051 -0.1169 -0.1308 -0.1459 -0.1616 -0.1686
3.0 15 Type A -0.1279 -0.1302 -0.1336 -0.1391 -0.1495 -0.1736 -0.1998
Type B -0.2408 -0.2666 -0.2986 -0.3361 -0.3772 -0.4198 -0.4385
30 Type A -0.0023 -0.0024 -0.0024 -0.0026 -0.0028 -0.0032 -0.0037
Type B -0.0957 -0.1058 -0.1182 -0.1327 -0.1487 -0.1652 -0.1725
205
4.6.1.5 Through-the-thickness variation of in-plane axial and shear
stresses for Type A FGSSP
In this sub section, variation of in-plane axial (xx and yy) and shear stresses (xy) are
studied for different thickness schemes of sandwich plates for linear variation of power law
exponent i.e., n=1.0. Different kinds of sandwich plates with thickness layup 2-1-2, 2-1-1,
1-1-1, 2-2-1, 1-2-1 are considered for the numerical problem. Plots showing variations of
in-plane axial stress xx for Type A plate corresponding to skew angles 15, 30, 45 and
60 are shown in Figure 4.34. As the skew angle of FGSSP deviates from 15 to 60,
magnitudes of stresses (both compressive and tensile) are significantly reduced. The stress
variation among the different types of FGSSP is reduced noticeably, when the skew angle
of the plate becomes 60. Next, through the thickness variation of in-plane normal stress
yy is also presented in Figure 4.35.
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
Thickness scheme
0.1 2-1-2 0.1 Thickness scheme
2-1-1 2-1-2
0.0 0.0
1-1-1 2-1-1
-0.1 2-2-1 -0.1 1-1-1
1-2-1 2-2-1
-0.2 -0.2
1-2-1
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-2 -1 0 1 2 3 4 5 6 7 -2 -1 0 1 2 3 4 5 6 7
In-plane normal stress (xx In-plane normal stress (xx
(a) Skew angle-15° 0.6 (b) Skew angle-30°
0.6
0.5
0.5
0.4 0.4
0.3 0.3
0.2 0.2
Thickness scheme
0.1 0.1
2-1-2
1-1-1 2-1-2
-0.1 -0.1
2-2-1 2-1-1
-0.2 1-2-1 -0.2 1-1-1
2-2-1
-0.3 -0.3
1-2-1
-0.4 -0.4
-0.5 -0.5
-1 0 1 2 3 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
In-plane normal stress (xx In-plane normal stress (xx
(c) Skew angle-45° (d) Skew angle-60°
Figure 4.34 Through-the-thickness variation of in-plane normal stress (xx) for different
thickness schemes of Type A FGSSP
206
Important observations drawn from Figure 4.34 regarding the type of stresses are
similar as in case of Figure 4.35. If skew angle value becomes 60, considerable variation
in magnitude of stresses is noticed, which is quite opposite to those observed in Figure
4.34. Variation of in-plane shear stress 𝜏xy through the thickness of the FGSSP is shown in
Figure 4.36. For FGSSP having skew angle 15, top of the plate exhibit compressive
stresses, while the bottom of the plate exhibit tensile stresses. For other values of skew
angles 30, 45 and 60, magnitude of stresses (both compressive and tensile) are not
significant. In addition, maximum stress values (in-plane normal and shear) for Type A
plate incorporating different ranges of skew angles and sandwich plate schemes are
presented in Table 4.53.
0.6 0.6
0.5 0.5
0.4 0.4
0.3 Non-dimensional depth (z/h)

0.3
0.2 0.2
0.1 0.1
Thickness scheme
2-1-1 2-1-2
-0.1 -0.1
1-1-1 2-1-1
-0.2 2-2-1 -0.2 1-1-1
1-2-1 2-2-1
-0.3 -0.3 1-2-1
-0.4 -0.4
-0.5 -0.5
-2 -1 0 1 2 3 4 5 6 7 -2 -1 0 1 2 3 4 5
In-plane normal stress (yy In-plane normal stress (yy
0.6 (a) Skew angle-15° 0.6

(b) Skew angle-30°
0.5 0.5
0.4 0.4
0.3
0.3
0.2 0.2
Thickness scheme
0.1 2-1-2 0.1
2-1-1 Thickness scheme
0.0
1-1-1 0.0 2-1-2
2-2-1 2-1-1
-0.1 -0.1
1-2-1 1-1-1
-0.2 -0.2 2-2-1
1-2-1
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-1 0 1 2 3 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0
In-plane normal stress (yy In-plane normal stress (yy
Figure 4.35 Through-the-thickness variation of in-plane normal stress (yy) for different
207
0.6
0.6
0.5
0.5
0.4
0.4

0.3
0.3
0.2 0.2
Thickness scheme Thickness scheme
0.1 2-1-2 0.1 2-1-2
2-1-1 0.0 2-1-1
0.0
1-1-1 1-1-1
-0.1 2-2-1 -0.1 2-2-1
1-2-1 1-2-1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-1.2 -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 -0.2 -0.1 0.0 0.1 0.2
In-plane normal stress (xy In-plane normal stress (xy
0.6 (a) Skew angle-15° 0.6 (b) Skew angle-30°
0.5
0.5
0.4
0.4
0.3
0.3

0.2
0.2
Thickness scheme 0.1
0.1 2-1-2
2-1-1 0.0 Thickness scheme
0.0 1-1-1 2-1-2
-0.1
-0.1 2-2-1 2-1-1
1-2-1 -0.2 1-1-1
-0.2 -0.3 2-2-1
1-2-1
-0.3 -0.4
-0.4 -0.5
-0.5 -0.6
-0.1 0.0 0.1 -0.1 0.0 0.1
In-plane normal stress (xy In-plane normal stress (xy
Figure 4.36 Through-the-thickness variation of in-plane shear stress (𝛕xy) for different
4.6.1.6 Through-the-thickness variation of in-plane axial and shear

stresses for Type B FGSSP
Non-dimensional quantities of in-plane axial (xx and yy) and shear stresses (xy) for Type
B FGSSP are plotted in Figures 4.36, 4.37 and 4.38. To perform the analysis, 1-8-1 skew
plate is chosen with a mixture of values of volume fraction index n= 1, 5 and 10. Identical
to Type A plate, the top of the plate experiences tensile stresses and bottom of the plate
experiences compressive stresses. Aluminium plate shows maximum compressive (at
bottom) and tensile stresses (at top) compared to Alumina and FGM plate. Once again, low
stiffness value of Aluminium plate is the reason for the above behavior. Also, further
increase in skew angle reduces the stresses (both compressive and tensile) in appreciable
manner as observed in the case of Type A plate. For FGSSP having 60 skew angle, there
is no significant change in the responses of different plates (both isotropic and FGM plate).
208
In Figure 4.38, for skew value 15, top (/bottom) of the plate records compressive (/tensile)
stresses for all the values of volume fraction indexes (n) considered.
Table 4.53: Non-dimensional values of in-plane normal and shear stresses for different values
of skew angles and thickness scheme for Type A FGSSP
Thickness  xx yy xy xz yz

scheme
2-1-2 15 4.9620 5.0832 0.8273 0.2069 0.2617
30 3.3261 3.6963 0.0263 0.0080 0.2385
45 1.6087 2.0942 0.0468 0.0980 0.1926
60 0.5002 0.8461 0.0112 0.0480 0.1282
2-1-1 15 5.8700 6.0131 0.9765 0.1755 0.2369

30 3.9326 4.3658 0.0299 0.0368 0.2173
45 1.8996 2.4593 0.0540 0.1195 0.1762
60 0.5869 0.9769 0.0121 0.0606 0.1171
1-1-1 15 5.2896 5.4188 0.8824 0.2001 0.2526

30 3.5459 3.9415 0.0283 0.0069 0.2302
45 1.7154 2.2356 0.0501 0.0936 0.1859
60 0.5339 0.9061 0.0121 0.0457 0.1237
2-2-1 15 5.9975 6.1439 0.9986 0.1810 0.2341

30 4.0193 4.4639 0.0310 0.0175 0.2138
45 1.9430 2.5209 0.0557 0.0987 0.1729
60 0.6025 1.0086 0.0129 0.0494 0.1150
1-2-1 15 6.0041 6.1509 1.0024 0.1927 0.2438

30 4.0253 4.4759 0.0326 0.0074 0.2222
45 1.9475 2.5431 0.0573 0.0908 0.1794
60 0.6068 1.0362 0.0142 0.0440 0.1195
0.6 0.6
0.5 0.5
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0.0 0.0 n=0

n=0 n=1
-0.1 -0.1 n=5
n=1
-0.2 n=5 -0.2 n=10
n=10 n=1000
-0.3 -0.3
n=1000
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
-0.8 -0.4 0.0 0.4 0.8 1.2 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8
(a) Skew angle-15° (b) Skew angle-30°
209
0.6 0.6
0.5 0.5

0.4 0.4
0.3 0.3
0.2 0.2
0.1 n=0 0.1
0.0 n=1 0.0 Ceramic

n=5 n=1.0
-0.1 n=10 -0.1
n=5.0
-0.2 n=1000 -0.2 n=10
Metal
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
-0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 -0.2 -0.1 0.0 0.1 0.2
thickness schemes of Type B FGSSP
0.6 0.6
0.4 0.4
0.2 0.2
0.0 n=0 0.0

n=0
n=1 n=1
n=5 n=5
-0.2 n=10 -0.2
n=10
n=1000 n=1000
-0.4 -0.4
-0.6 -0.6
-0.8 -0.4 0.0 0.4 0.8 1.2 -0.8 -0.4 0.0 0.4 0.8
0.6
(a) Skew angle-15° 0.6
(b) Skew angle-30°
0.4 0.4
0.2 0.2
0.0 0.0 n=0

n=0
z/h
n=1 n=1
n=5 n=5
-0.2 -0.2 n=10
n=10
n=1000 n=1000
-0.4 -0.4
-0.6 -0.6
-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 -0.2 0.0 0.2
210
0.6
0.6
0.4
0.4

0.2
0.2
0.0 n=0
n=1 0.0
n=0
n=5
n=1
-0.2 n=10 n=5
n=1000 -0.2
n=10
n=1000
-0.4
-0.4
-0.6 -0.6
-0.2 0.0 0.2 -0.2 -0.1 0.0 0.1 0.2
0.6 (a) Skew angle-15° 0.6 (b) Skew angle-30°
0.4 0.4
0.2 0.2
0.0 n=0 0.0 n=0

n=1 n=1
n=5 n=5
-0.2 n=10 -0.2 n=10
n=1000 n=1000
-0.4 -0.4
-0.6 -0.6
-0.2 -0.1 0.0 0.1 0.2 -0.2 -0.1 0.0 0.1 0.2
Also, metal plate gives the maximum bending stresses (both compressive and
tensile) in the plate. As the value of skew value of FGSSP increases, no significant changes
in the stresses are observed irrespective of the type of thickness scheme of FGSSP.
Variations of in-plane axial stress and shear stress for Type B FGSSP are also shown in
Table 4.54. As observed from previous cases as the skew angle increases, magnitude of
stresses (in-plane normal and shear stresses) reduces. Also, for particular value of skew
angle, as the volume fraction index increases in-plane normal and shear stresses increases.
Hence it can be inferred that two parameters i.e., skew angle and volume fraction index
play vital role in predicting the magnitude of stresses. Due to this observation an additional
care has to be exercised in choosing these parameters.
211
Table 4.54: Non-dimensional in-plane normal and shear stresses for different values of skew
angles and thickness scheme for Type B FGSSP
n  xx yy xy xz yz

Ceramic 15 0.4406 0.4514 0.0737 0.0384 0.0451
30 0.2956 0.3288 0.0025 0.0053 0.0408
45 0.1434 0.1877 0.0043 0.0101 0.0328
60 0.0452 0.0780 0.0011 0.0047 0.0219
1.0 15 0.5426 0.5558 0.0908 0.0659 0.0727

30 0.3642 0.4051 0.0031 0.0182 0.0653
45 0.1770 0.2316 0.0053 0.0066 0.0523
60 0.0561 0.0968 0.0014 0.0025 0.0349
5.0 15 0.6649 0.6811 0.1117 0.1823 0.1687

30 0.4482 0.4984 0.0040 0.1144 0.1480
45 0.2204 0.2885 0.0069 0.0571 0.1166
60 0.0728 0.1250 0.0022 0.0313 0.0782
10.0 15 0.7922 0.8116 0.1328 0.2152 0.1973

30 0.5338 0.5931 0.0046 0.1384 0.1729
45 0.2624 0.3418 0.0081 0.0714 0.1360
60 0.0865 0.1464 0.0025 0.0389 0.0913
Metal 15 1.0775 1.1037 0.1799 0.2366 0.2201

30 0.7247 0.8044 0.0059 0.1460 0.1932
45 0.3547 0.4593 0.0106 0.0713 0.1523
60 0.1152 0.1915 0.0029 0.0393 0.1023
4.6.2 Static analysis of FGM sandwich skew shells subjected to

mechanical loading
To perform bending analysis of functionally graded sandwich skew shells (FGSSS), a
simply supported FGM shell panel consists of three elastic inhomogeneous layers
x  y 
subjected to sinusoidal load of intensity q  x, y   q0 sin   sin   , where q0 in is the
 a   a 
intensity of load at the plate center is considered. In this study, the ceramic content was
chosen to be Zirconia (ZrO2), along with aluminium (Al) as metallic component of
sandwich shell panel. The volume fraction distribution of constituents of shell panel
follows power law distribution, in the manner that ceramic content attains unity at the top
of the shell and approaches zero at the bottom of the shell panel.
212
4.6.2.1 Validation of results of static analysis of FGM sandwich skew
shells subjected to mechanical loading
Since no results are available for bending analysis of FGSSS, the case of functionally
graded shell based on FSDT in conjunction with element free Ritz method (Zhao et al.
2009c) are taken for validation purpose. The deflection results are tabulated (Table 4.55)
for simply supported and clamped boundary with the thickness value ranges from 0.02m to
0.005m. It is noticed that the deviation among the third order and FSDT corresponds to
simply supported case is more than clamped boundary. The present theory considers the
parabolic shear variation, wherein the linear variation is considered in case of FSDT. In
addition, the solution methodologies proposed by the authors may be the probable reason
for the deviation. Convergence study was performed for Aluminium/Zirconia shell and it
was observed that 6x6 mesh division satisfied the convergence criteria. Hence for the
subsequent analyses 6x6 mesh division was considered and the interesting and important
observations were made with respect to bending and stress behavior of FGSSS.
Table 4.55: Deflection of Al/ZrO2 shell for various radius-thickness ratios (R/h) and volume
fraction indexes (n)
SSSS CCCC
R/h Reference n=0.5 n =1.0 n =2.0 n =0.5 n =1.0 n =2.0
50 Zhao et al. (2009c) 0.0038 0.0042 0.0046 0.0013 0.0014 0.0016

Present 0.0034 0.0037 0.0040 0.0013 0.0014 0.0016
% of error 10.52 11.90 13.04 0 0 0
100 Zhao et al. (2009c) 0.0542 0.0607 0.0665 0.0171 0.0191 0.0210
Present 0.0393 0.0428 0.0467 0.0164 0.0185 0.0204
% of error 27.49 29.48 29.77 4.09 3.14 2.86
200 Zhao et al. (2009c) 0.6503 0.7283 0.8057 0.2026 0.2269 0.2509
present 0.5293 0.6601 0.7990 0.1938 0.2184 0.2416
% of error 18.60 9.36 0.83 4.34 3.74 3.70
4.6.2.2 Influence of side-thickness ratio (a/h) on non-dimensional central

deflection of Type A and Type B FGSSS
Non-dimensional transverse displacements for three kinds of Type A FGSSS (thickness
scheme 1-1-1, 1-2-1 and 2-1-2), for several values of gradient parameter (n=0, 0.5, 1, 2, 5
and 10.0) and for different values of a/h (5, 10, 20, 30, 50 and 100) are shown in Figure
4.40. A sharp rise in the displacement is manifested for all the kinds of sandwich shells and
this trend is general for all the values of gradient parameter under consideration. As
expected, increase in gradient parameter tends the shell panel to deflect more owing to the
213
low stiffness property attributed by the metal component. Regarding the influence of
thickness ratio (a/h) much deviation was not recorded by the homogenous and graded shell
up to the range of a/h=10 (i.e., moderately thick shell). The divergence becomes
significant when the shell enters into the zone of thin category. For instance, for a/h=100
(thin shell), homogenous and graded shells have shown marked divergence of
displacement for different thickness schemes of FGSSS.
0.0016 0.0016
n=0
n=0.2

n=0 n=0.5
n=0.2 n=1
0.0012 0.0012
n=0.5 n=2
n=1 n=5
n=2 n=10
n=5
0.0008 n=10 0.0008
0.0004 0.0004
0.0000 0.0000
0 20 40 60 80 100 0 20 40 60 80 100
Side-thickness ratio (a/h) Side-thickness ratio (a/h)

0.0016
n=0
n=0.2
n=0.5
n=1
0.0012
n=2
n=5
n=10
0.0008
0.0004
0.0000
0 20 40 60 80 100
Side-thickness ratio (a/h)
Figure 4.40 Influence of thickness ratio (a/h) on non-dimensional central deflection of

Al/ZrO2 Type A FGSSS
The maximum displacement shown in Figure 4.41 for Type B FGSSS having 1-1-1, 1-2-
1, 2-1-2 thickness range exhibits the increase in tendency and attains the peak of the
component as the shell approaches the value of a/h=100. Among the several values of
gradient parameter considered, ceramic shell i.e., n=0 gives minimum displacement
provided the volume fraction distribution follows power law formula. High bending
214
stiffness offered by the homogenous ceramic shell is the well known cause for this
tendency.
0.0012
0.0014
0.0010 0.0012

0.0010
0.0008
0.0008
0.0006 n=0 n=0
n=0.2 n=0.2
0.0006 n=0.5
n=0.5
n=1
0.0004 n=1
n=2
n=2 0.0004
n=5
n=5
n=10
0.0002 n=10
0.0002
0.0000 0.0000
0 20 40 60 80 100 0 20 40 60 80 100
(a) 1-1-1 thickness scheme Side-thickness ratio (a/h)
0.0012
(b) 1-2-1 thickness scheme
0.0010
0.0008
n=0
0.0006
n=0.2
n=0.5
n=1
0.0004
n=2
n=5
n=10
0.0002
0.0000
0 20 40 60 80 100
Figure 4.41 Influence of thickness ratio (a/h) on non-dimensional central deflection of

Al/ZrO2 Type B FGSSS
Again, different values of gradient parameter will not show any discernible influence on
maximum displacement when the shell fall under moderately thick category, which is
similar to the observation recorded for Type A FGSSS. In Type B FGSSS, the deflection
curves are closely spaced compared to Type A FGSSS, and this observation is more
pronounced when the thickness of the skin layers are twice than thickness of core portion
(refer Figure 4.41.(c)). This is evident, since the behavior of homogenous shell is
independent of the gradient parameter unlike graded shells.
215
4.6.2.3 Influence of radius-side ratio (R/a) on non-dimensional central
deflection of Type A and Type B FGSSS
0.00060 0.00060

0.00054 0.00054
n=0 n=0
0.00048 n=0.2 0.00048 n=0.2
n=0.5 n=0.5
n=1 n=1
0.00042 0.00042 n=2
n=2
n=5 n=5
n=10 n=10
0.00036 0.00036
0.00030 0.00030
0 20 40 60 80 100 0 20 40 60 80 100
Curvature-side ratio (Rx/a) Curvature-side ratio (Rx/a)
0.00066
0.00060
n=0
0.00054 n=0.2
n=0.5
n=1
0.00048
n=2
n=5
0.00042 n=10
0.00036
0.00030
0 20 40 60 80 100
Curvature-side ratio (Rx/a)
Figure 4.42 Influence of curvature-side ratio (R/a) on non-dimensional central deflection of

Al/ZrO2 Type A FGSSS
The highest magnitude of the maximum displacements for different values of curvature
(Rx=1/Rx) is observed for 2-1-2 thickness scheme for Type A FGSSS and it is shown in
Figure 4.42. This is due to the high thickness proportion of graded shell with those
compared to isotropic shell. Sudden rise is observed for the transition of curvatures from
5.0 to 10.0 and the point at which this occurs represents the portion of FGM shell. The
displacement component becomes invariant when the value of (R/a) increases above 20.
When similar study was performed for Type B FGSSS (Figure 4.43), 2-1-2 type FGSSS
gives more displacement for homogenous and graded shell category once again. Initially
displacement elevates sharply and reduces further beyond the value Rx/a=20, and this
216
reduction is more visible when core portion thickness is more compared to homogenous
top and bottom segment. However, 1-2-1 type sandwich shell records minimum deflection
in the group.
0.000450
0.000456
0.000448 0.000435

0.000440
0.000420
0.000432
0.000405 n=0
n=0 n=0.2
0.000424
n=0.2 n=0.5
n=0.5 0.000390
n=1
0.000416 n=1 n=2
n=2 n=5
n=5 0.000375
0.000408 n=10
n=10
0.000360
0 20 40 60 80 100 0 20 40 60 80 100
Curvature-side ratio (Rx/a) Curvature-side ratio (Rx/a)
0.00046
0.00045
0.00044
n=0
n=0.2
n=0.5
n=1
n=2
n=5
0.00043
n=10
0 20 40 60 80 100
Curvature-side ratio (Rx/a)
Figure 4.43 Influence of curvature-side ratio (R/a) on non-dimensional central deflection of

Al/ZrO2 Type B FGSSS
4.6.2.4 Variation of non-dimensional in-plane normal and shear stresses

for Type A and Type B Al/ZrO2 FGSSS
Figure 4.44 demonstrates the in-plane stress (σxx) distribution for Type A FGSSS for
gradient parameter n=1.0, 5.0 and 10.0. Transition from tensile stress distribution to
compression occurs at the reference axis of the shell. Such transition becomes smooth as
the thickness of core portion is reduced with respect to top and bottom face layers.
Maximum magnitude of tensile and compression are produced by 2-1-2 FGSSS and this
fact is observed to be common for different values of gradient parameter considered. For
the case of n=10.0, the linear transition is observed by the 2-1-2 FGSSS.
217
0.6 0.6
0.5 0.5

0.4
0.4
0.3 0.3
0.2 0.2 Thickness scheme

Thickness scheme
0.1 2-1-2 0.1 2-1-2
1-1-1 1-1-1
0.0 0.0
1-2-1 1-2-1
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
-400000 0 400000 800000 -400000 -200000 0 200000 400000
(a) Volume fraction index (n)=1.0 (b) Volume fraction index (n)=5.0
thickness schemes of Type A FGSSS
0.6
0.6
0.5 0.5
0.4
0.4
0.3 0.3
Thickness scheme
Thickness scheme
2-1-2
0.2 2-1-2 0.2 1-1-1
1-1-1 1-2-1
0.1 0.1
1-2-1
0.0 0.0
-0.1 -0.1
-0.2 -0.2
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
-400000 0 400000 800000 -200000 0 200000
(a) Volume fraction index (n)=1.0 0.6 (b) Volume fraction index (n)=5.0
0.6
0.5
0.5
0.4
0.4
0.3
0.3 Thickness scheme Thickness scheme
0.2
2-1-2 0.2 2-1-2
1-1-1 0.1 1-1-1
0.1 1-2-1 1-2-1
0.0
0.0
-0.1
-0.1
-0.2
-0.2
-0.3
-0.3
-0.4
-0.4
-0.5
-0.5
-0.6
-0.6
-200000 -100000 0 100000 200000
-400000 0 400000
In-plane normal stress (yy) In-plane normal stress (xx)
(c) Volume fraction index (n)=10 (c) Volume fraction index (n)=10.0
218
Exception case with compression and tensile stresses at the top and bottom, respectively,
was noticed by 1-2-1 FGSSS with the gradient parameter approaches the value n=10.0.
Further magnitudes of stresses are reduced as the value of gradient parameter rises. For
Type A FGSSS, as the n value increases volume fraction of ceramic tends to reduce and
the volume fraction of metal increases.
0.5 0.6
0.4 0.5
0.4

0.3
0.3
0.2
0.2
0.1 Thickness scheme Thickness scheme
0.1
2-1-2 2-1-2
0.0 1-1-1 0.0 1-1-1
1-2-1 -0.1 1-2-1
-0.1
-0.2
-0.2
-0.3
-0.3
-0.4
-0.4 -0.5
-0.5 -0.6
-800000 -400000 0 400000 800000 -400000 -200000 0 200000 400000
In-plane normal stress (xy) In-plane normal stress (xy)
(a) Volume fraction index (n)=1 (b) Volume fraction index (n)=5
0.6
0.5
0.4
0.3
0.2
Thickness scheme
0.1
2-1-2
0.0 1-1-1
1-2-1
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-200000 -100000 0 100000 200000
In-plane normal stress (xy)
(c) Volume fraction index (n)=10
Again this observation is dependent on type of thickness scheme adopted for the modeling.
Through the thickness variation of σyy depicted in Figure 4.45 for the value of n=1.0, 5.0
and 10.0, respectively, shows the stress pattern similar to Figure 4.44. In plot, 2-1-2 Type
FGSSS records maximum stresses for the different values of gradient parameter (n) under
consideration. It is interesting to observe from Fig.4.45 (c) that, when FGSSS was modeled
to have larger core portion of homogenous material compared to face sheets (i.e., 1-2-1
219
type), the stress distribution is insignificant. Thus it is put forward that 1-2-1 sandwich
shell suppresses any tendency of cracks to grow in both tensile and compressive mode, and
this tendency is pronounced when the value of n chosen is 10.0.
Through-the-thickness variations of in-plane shear stress (τxy) for 1-1-1, 1-2-1 and
2-1-2 FGSSS with the gradient parameter n, shown in Figure 4.46 were found to produce
compressive stresses at the upper half zone while tensile stresses are observed in the
bottom zone of the shell. This stress pattern was identical for different volume fractions of
ceramic and metal and high magnitude of stresses were obtained corresponding to low
value of gradient parameter. As expected, 2-1-2 sandwich type produce maximum stresses
(both compression and tension), exception was found near at the top and bottom of the
shell for the case of n=10.0 (refer Figure 4.45 (c)). It is noted that these portions are
represented by 100% volume fraction of metal component.
0.6
0.6
0.5
0.5
0.4
0.4
0.3 0.3
0.2 0.2
0.1 0.1
Thickness scheme
0.0 0.0
2-1-2 Thickness scheme
-0.1 1-1-1 -0.1 2-1-2
1-2-1 1-1-1
-0.2 -0.2
1-2-1
-0.3 -0.3
-0.4 -0.4
-0.5 -0.5
-0.6 -0.6
-1000000 0 1000000 2000000 3000000 4000000 -1000000 0 1000000 2000000 3000000 4000000
(a) In-plane normal stress (xx) (b) In-plane normal stress (yy)
0.6
0.5
0.4
0.3
0.2
0.1
Thickness scheme
0.0 2-1-2
1-1-1
-0.1
1-2-1
-0.2
-0.3
-0.4
-0.5
-0.6
-4000000 -2000000 0 2000000
(c) In-plane shear stress (xy)
Figure 4.47 Through-the-thickness variation of in-plane normal (xx and yy) and shear
stresses (𝛕xy) for different thickness schemes of Type B FGSSS (n=1)
220
In-plane stress variations (σxx and σyy) for type B FGSSS for linear variation of gradient
parameter is demonstrated Figure 4.46. Maximum compressive stresses are observed at
metal rich segment for all the kinds of FGSSS. Maximum tensile stresses are noticed at the
ceramic rich top segment corresponding to 2-1-2 thickness scheme. Symmetric FGSSS
(i.e., thickness scheme 1-1-1) shows intermediate stress distribution for the different value
of gradient parameter. Similar observations regarding stress pattern and type of sandwich
scheme is obvious for in-pane shear stress (τxy), except the nature of stresses are different
at top and bottom segment (Figure 4.46 (c)).
4.7 FREE VIBRATION ANALYSIS OF FGM SANDWICH SKEW

PLATES AND SHELLS
To perform the vibration study for FGM sandwich plates and shells, four kinds of
thickness schemes are considered. Under this section, 1-1-1, 2-1-2, 1-2-1, and 1-8-1
thickness schemes are represented as Type 1, Type 2, Type and Type 4 plate/shell,
respectively. The non-dimensional frequency parameter used for the free vibration study of
sandwich skew plate is given below.
a2 0
Non-dimensional frequency:  
h E0
where ρ0 and E0 are the reference values of density and Young’s modulus and taken as
ρ0=1 kg/m3 and E0=1GPa.
4.7.1 Validation of free vibration results of FGSSP
In this section, free vibration problem of FGSP has been validated with the results
available in the literatures. Once the close agreement between the results is ensured, new
results are generated for FGSSP and FGSSS by considering different values of skew
angles.
For validation, a simply supported FGM plate comprising of Aluminium/Alumina
constituents is considered for various values of side-thickness ratios (a/h). In Table.4.56, a
close agreement among the vibration results is observed. Also, percentage difference
between the present results and results by Natarajan and Ganapathi (2012) using HSDT 13
model is shown in Table 4.56.
221
Table 4.56: Validation study for free vibration of FGSP for different thickness and volume
fraction index values
n
a/h References
0 0.5 1 5
5 Natarajan and Ganapathi (2012) HSDT 13 1.6774 1.4219 1.2778 0.9986
Natarajan and Ganapathi (2012) HSDT 11 1.6774 1.4219 1.2778 0.9988
Natarajan and Ganapathi (2012) FSDT 1.6689 1.4076 1.2628 0.9860
Li et al. (2008) 1.6771 1.4218 1.2777 0.9980
Present (16X16)a 1.5783 1.3561 1.2211 0.9711
% of error 5.90 4.62 4.43 2.75

Li et al. (2008) 1.8268 1.5213 1.3552 1.0453
Present (16X16) 1.8339 1.5321 1.3598 1.0444
% of error 0.38 0.70 0.33 0.10
Li et al. (2008) 1.8883 1.5605 1.3851 1.0631
Present (16X16) 1.9211 1.5820 1.4073 1.0929
% of error 1.73 1.37 1.59 2.80
4.7.1.1 Influence of boundary conditions on frequency response of FGSP

having graded and homogeneous core
Table 4.57: Influence of boundary conditions on frequency parameter of FGSSP having
graded core (=15°)
Types Boundary n
conditions Ceramic 2 5 10 Metal
Type 1 SSSS 1.9563 1.3980 1.2246 1.1625 1.1070
CCCC 1.1070 1.1070 1.1070 1.1070 1.1070
SCSC 1.1070 1.1070 1.1070 1.1070 1.1070
CFCF 1.1070 1.1070 1.1070 1.1070 1.1070
Type 2 SSSS 1.9598 1.3061 1.1363 1.0872 1.0503

CCCC 1.0503 1.0503 1.0503 1.0503 1.0503
SCSC 1.0503 1.940 1.6879 1.6113 1.5511
CFCF 2.2179 1.4739 1.2777 1.2227 1.1827
Type 3 SSSS 1.9414 1.5306 1.3683 1.2988 1.2296

CCCC 3.3186 2.6121 2.3125 2.1804 2.0463
SCSC 2.7782 2.211 1.9683 1.8592 1.7471
CFCF 2.1797 1.6864 1.4868 1.3996 1.3114
Type 4 SSSS 1.8516 1.8009 1.7111 1.6651 1.6167

CCCC 3.0187 2.8475 2.6678 2.5752 2.4769
SCSC 2.5395 2.420 2.2770 2.2033 2.1247
CFCF 1.9703 1.8338 1.7111 1.6483 1.5817
222
Types Boundary n
Type 1 SSSS 2.4474 1.8077 1.5994 1.5206 1.4472
CCCC 1.4472 1.4472 1.4472 1.4472 1.4472
SCSC 1.4472 1.4472 1.4472 1.4472 1.4472
CFCF 1.4472 1.4472 1.3724 1.2986 1.2325
Type 2 SSSS 2.4528 1.6954 1.4866 1.4195 1.3642

CCCC 1.3642 1.3642 1.3642 1.3642 1.3642
SCSC 3.3793 2.3518 2.0535 1.9592 1.8826
CFCF 2.2256 1.4820 1.2853 1.2298 1.1891
Type 3 SSSS 2.4250 1.9580 1.7643 1.6795 1.5932

CCCC 1.5932 1.5932 1.5932 1.5932 1.5932
SCSC 3.3208 2.6783 2.3965 2.2697 2.1390
CFCF 2.1872 1.6957 1.4963 1.4092 1.3211
Type 4 SSSS 2.3002 2.2523 2.1440 2.0884 2.0296

CCCC 2.0296 2.0296 2.0296 2.0296 2.0296
SCSC 2.0296 2.0296 2.0296 2.6942 2.6041
CFCF 1.9784 1.8446 1.7222 1.6596 1.5934

Types Boundary n
Type 1 SSSS 3.5496 2.6946 2.3982 2.2801 2.1666
CCCC 2.1666 2.1666 2.1666 2.1666 2.1666
SCSC 2.1666 2.1666 3.1347 2.9752 2.8243
CFCF 2.2782 1.6316 1.4222 1.3463 1.2780
Type 2 SSSS 3.5589 2.5319 2.2281 2.1234 2.0322

CCCC 2.0322 2.0322 2.0322 2.0322 2.0322
SCSC 4.6890 3.3327 2.9224 2.7867 2.6723
CFCF 2.2853 1.5328 1.3312 1.2732 1.2300
Type 3 SSSS 3.5129 2.8998 2.6299 2.5081 2.3821

CCCC 2.3821 2.3821 2.3821 2.3821 2.3821
SCSC 4.6135 3.7923 3.4168 3.2463 3.0701
CFCF 2.2465 1.7528 1.5503 1.4616 1.3718
Type 4 SSSS 3.3254 3.2844 3.1342 3.0568 2.9747

CCCC 2.9747 2.9747 2.9747 2.9747 2.9747
SCSC 2.9747 2.9747 2.9747 3.8706 3.7527
CFCF 2.0360 1.9066 1.7831 1.7199 1.6529
Under this section, an attempt has been exercised to tabulate the frequency results for
Aluminium/Alumina sandwich plate (h=0.1m). To illustrate the influence of boundary
conditions four kinds of boundary conditions, namely, simply supported, clamped and free
are considered. Similarly, different values of skew angles ranging from 15° to 60° are also
considered to execute the vibration analysis.
223
Types Boundary n
Type 1 SSSS 6.2862 4.8928 4.3721 4.1554 3.9419
CCCC 3.9419 3.9419 3.9419 3.9419 3.9419
SCSC 7.9815 6.2409 5.5612 5.2861 5.0220
CFCF 2.4342 1.7700 1.5507 1.4697 1.3958
Type 2 SSSS 6.3051 4.6022 4.0568 3.8595 3.6819

CCCC 9.4448 6.9871 6.1449 5.8588 5.6137
SCSC 8.0055 5.8752 5.1747 4.9301 4.7161
CFCF 2.4415 1.6631 1.4492 1.3848 1.3349
Type 3 SSSS 6.2164 5.2489 4.7893 4.5760 4.3515

CCCC 4.3515 4.3515 4.3515 4.3515 4.3515
SCSC 4.3515 4.3515 6.0697 5.7887 5.4954
CFCF 2.4017 1.9004 1.6907 1.5983 1.5045
Type 4 SSSS 5.9045 5.9069 5.6575 5.5284 5.3909

CCCC 5.3909 5.3909 5.3909 5.3909 5.3909
SCSC 5.3909 5.3909 7.1212 6.9515 6.7699
CFCF 2.1859 2.0674 1.9412 1.8766 1.8083

homogeneous core (=15°)
Types Boundary n
Type 1 SSSS 1.5098 1.4684 1.4317 1.3948 1.3397
CCCC 1.3397 2.4309 2.3266 2.2248 2.0767
SCSC 2.0767 2.0591 1.9727 1.8889 1.7665
CFCF 1.6517 1.5817 1.5170 1.4514 1.3555
Type 2 SSSS 1.4480 1.4190 1.3970 1.3775 1.3513

CCCC 1.3513 1.3513 1.3513 1.3513 1.3513
SCSC 1.3513 1.3513 1.3513 1.3513 1.3513
CFCF 1.3513 1.3513 1.3513 1.3513 1.3513
Type 3 SSSS 1.5835 1.5218 1.4650 1.4075 1.3156

CCCC 1.3156 1.3156 2.4265 2.2878 2.0597
SCSC 2.3036 2.1630 2.0427 1.9295 1.7440
CFCF 1.7440 1.6978 1.6107 1.5179 1.3622
Type 4 SSSS 1.6908 1.5403 1.4128 1.3152 1.1636

CCCC 1.1636 1.1636 1.1636 2.1209 1.9130
SCSC 2.3552 2.0783 1.8787 1.7730 1.5962
CFCF 1.8344 1.6607 1.5322 1.4507 1.2918
Frequency values for FGSSS having graded core is presented in Tables 4.57-4.60; while
for FGSSS having homogenous core, frequency results are shown in Tables 4.61-4.64. As
the skew angle increases, the frequency value increases for all the kinds of sandwich plates
considered. If skew angle increases, stiffness of the plate increases and thus the
corresponding frequency values are increased. Among this group, Type 2 FGSSP ensures
high frequency values for all skew angles. While for other cases (Type1, Type 3 and Type
224
4 FGSSP) records close range of frequency values. Due to stiffness degradation, increase
in volume fraction index records lower deflection values for different boundary conditions
and skew angles assumed in this example. Further, for clamped boundary condition at all
the edges, high stiffness gives the maximum value of free vibration frequency.
Types Boundary n
Type 1 SSSS 1.9082 1.8340 1.7758 1.7212 1.6420
CCCC 1.6420 1.6420 1.6420 1.6420 1.6420
SCSC 1.6420 1.6420 1.6420 1.6420 2.1220
CFCF 1.6600 1.5898 1.5254 1.4602 1.3648
Type 2 SSSS 1.8203 1.7697 1.7343 1.7042 1.6648

CCCC 1.6648 1.6648 1.6648 1.6648 1.6648
SCSC 1.6648 1.6648 1.6648 1.6648 1.6648
CFCF 1.5370 1.4852 1.4461 1.4116 1.3665
Type 3 SSSS 2.0004 1.8920 1.8064 1.7277 1.6035

CCCC 1.6035 1.6035 1.6035 1.6035 1.6035
SCSC 1.6035 1.6035 1.6035 1.6035 2.0806
CFCF 1.7931 1.7045 1.6176 1.5254 1.3711
Type 4 SSSS 2.1098 1.8829 1.7108 1.5951 1.4186

CCCC 1.4186 1.4186 1.4186 1.4186 1.4186
SCSC 1.4186 2.4772 2.2238 2.0943 1.8884
CFCF 1.8419 1.6661 1.5360 1.4537 1.2956

Types Boundary n
Type 1 SSSS 2.7987 2.6650 2.5654 2.4758 2.3473
CCCC 2.3473 2.3473 2.3473 2.3473 2.3473
SCSC 3.6542 3.4568 3.3038 3.1648 2.9653
CFCF 1.7102 1.6349 1.5674 1.4999 1.4017
Type 2 SSSS 2.6640 2.5743 2.5132 2.4624 2.3964

CCCC 2.3964 2.3964 2.3964 2.3964 2.3964
SCSC 3.4394 3.3046 3.2108 3.1325 3.0306
CFCF 1.5847 1.5297 1.4888 1.4529 1.4061
Type 3 SSSS 2.9265 2.7278 2.5832 2.4608 2.2705

CCCC 2.2705 2.2705 2.2705 2.2705 2.2705
SCSC 2.2705 2.2705 2.2705 2.2705 2.8722
CFCF 1.8452 1.7492 1.6579 1.5627 1.4044
Type 4 SSSS 3.0551 2.6658 2.3886 2.2293 1.9963

CCCC 1.9963 1.9963 1.9963 1.9963 1.9963
SCSC 1.9963 1.9963 3.0207 2.8379 2.5704
CFCF 1.8960 1.7068 1.5699 1.4850 1.3235
225
Types Boundary n
Type 1 SSSS 5.0238 4.7462 4.5448 4.3700 4.1190
CCCC 4.1190 4.1190 4.1190 4.1190 4.1190
SCSC 6.3652 5.9709 5.6836 5.4365 5.0851
CFCF 1.8393 1.7521 1.6773 1.6044 1.4995
Type 2 SSSS 4.7927 4.6081 4.4835 4.3813 4.2478
CCCC 4.2478 4.2478 4.2478 4.2478 4.2478
SCSC 6.0384 5.7748 5.5974 5.4535 5.2674
CFCF 1.7077 1.6450 1.5997 1.5607 1.5102
Type 3 SSSS 5.2211 4.7946 4.5008 4.2728 3.9234
CCCC 3.9234 3.9234 3.9234 3.9234 3.9234
SCSC 6.6408 6.0390 5.6251 5.3102 4.8269
CFCF 1.9805 1.8679 1.7658 1.6617 1.4927
Type 4 SSSS 5.4192 4.5949 4.0316 3.7660 3.4074
CCCC 3.4074 3.4074 3.4074 3.4074 3.4074
SCSC 3.4074 3.4074 4.9358 4.6284 4.2314
CFCF 2.0365 1.8184 1.6671 1.5750 1.3782
4.7.1.2 Influence of thickness-side ratio on frequency parameter of

FGSSP
Table 4.65: Influence of thickness-side ratio (h/b) on non-dimensional frequency values of
FGSSP having graded core for different volume fraction index and skew angle values (Type 1)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.5961 1.0168 1.0168 1.0147 1.0147 1.0047 1.0147 1.0147
0.1 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147
0.2 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147
30° 0.01 1.0788 1.0788 1.0788 1.1199 1.1199 1.1199 1.1199 1.1199
0.1 1.1199 1.1199 1.1199 1.1199 1.1199 1.1199 1.1199 1.1199
0.2 1.1199 1.1199 1.1199 1.1199 1.1199 1.1199 1.1199 1.1199
45° 0.01 1.0291 1.0291 1.0291 1.0291 1.0291 1.0291 1.0291 1.0291
0.1 1.0291 1.0291 1.0291 1.0291 1.0291 1.0291 1.0291 1.0291
0.2 1.0291 1.0291 1.0291 1.0291 1.0290 1.0291 1.0291 1.0291
60° 0.01 2.5456 2.0233 1.5747 1.1500 1.1500 1.1500 1.1426 1.1989
0.1 1.1989 1.1989 1.1989 1.1989 1.1989 1.1989 1.1989 1.1989
0.2 1.1989 1.1989 1.1989 1.1989 1.1989 1.1989 1.1989 1.1989

n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.9009 1.0014 1.0014 1.0014 1.0014 1.0014 1.0014 1.0014
0.1 1.0014 1.0014 1.0014 1.0014 1.0014 1.0014 1.0014 1.0014
0.2 1.0014 1.0014 1.0014 1.0014 1.0014 1.0014 1.0014 1.0014
30° 0.01 3.0340 1.9657 1.3158 1.0591 1.0964 1.0418 1.0418 1.0418
0.1 1.0418 1.0418 1.0418 1.0418 1.0418 1.0418 1.0418 1.0418
0.2 1.0418 1.0418 1.0418 1.0418 1.0418 1.0418 1.0418 1.0418
45° 0.01 8.0796 1.5285 1.5285 1.5285 1.0147 1.0147 2.9846 1.1107
0.1 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107
0.2 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107
60° 0.01 1.2937 1.2937 1.1991 1.1991 1.1063 1.1063 1.1063 1.1063
0.1 1.1063 1.1063 1.1063 1.1063 1.1063 1.1063 1.1063 1.1063
0.2 1.1063 1.1063 1.1063 1.1063 1.1063 1.1063 1.1063 1.1063
226
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.4572 1.3180 1.1334 1.1869 1.0794 1.0794 1.0794 1.0470
0.1 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470
0.2 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470
30° 0.01 1.1432 1.0522 1.0185 1.0185 1.0836 1.0836 1.0836 1.0836
0.1 1.0836 1.0836 1.0836 1.0836 1.0836 1.0836 1.0836 1.0836
0.2 1.0836 1.0836 1.0836 1.0836 1.0836 1.0836 1.0836 1.0836
45° 0.01 1.9763 1.0408 1.0408 1.0408 1.3790 1.1101 1.0254 1.1650
0.1 1.1650 1.1650 1.1650 1.1650 1.1650 1.1650 1.1650 1.1650
0.2 1.1650 1.1650 1.1650 1.1650 1.1650 1.1650 1.1650 1.1650
60° 0.01 1.0208 1.0010 1.0366 1.0366 1.0366 1.0366 1.0366 1.0366
0.1 1.0366 1.0366 1.0366 1.0366 1.0366 1.0366 1.0366 1.0366
0.2 1.0366 1.0366 1.0366 1.0366 1.0366 1.0366 1.0366 1.0366

n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.1122 1.0741 1.0949 1.0969 1.0746 1.0746 1.0746 1.0746
0.1 1.0746 1.0746 1.0746 1.0746 1.0746 1.0746 1.0746 1.0746
0.2 1.0746 1.0746 1.0746 1.0746 1.0746 1.0746 1.0746 1.0746
30° 0.01 1.4965 1.1099 1.2920 1.0031 1.0031 1.0031 1.0494 1.0494
0.1 1.0494 1.0494 1.0494 1.0494 1.0494 1.0494 1.0494 1.0494
0.2 1.0494 1.0494 1.0494 1.0494 1.0494 1.0494 1.0494 1.0494
45° 0.01 1.1696 1.2836 1.3887 1.1552 1.1552 1.3100 1.3100 1.0570
0.1 1.0570 1.0570 1.0570 1.0570 1.0570 1.0570 1.0570 1.0570
0.2 1.0570 1.0570 1.0570 1.0570 1.0570 1.0570 1.0570 1.0570
60° 0.01 1.3161 1.3161 1.0901 1.0901 1.0901 1.0760 1.0475 1.0663
0.1 1.0663 1.0663 1.0663 1.0663 1.0663 1.0663 1.0663 1.0663
0.2 1.0663 1.0663 1.0663 1.0663 1.0663 1.0663 1.0663 1.0663
Influence of thickness on fundamental frequency of FGSSP is studied for different kinds of

sandwich schemes in Tables 4.65-4.68 for graded core and in Tables 4.69-4.72 for
homogeneous core, for different values of skew angles. For thickness-side ratio (h/b)
values 0.1 m and 0.2 m, equal value of frequency parameter is observed for all skew angles
and sandwich schemes. But for thin plate cases (h=0.01), quite different results are
observed for all the types of FGSSP. No stable effect of volume fraction index on
frequency values is observed in all cases. In some cases, constant value of frequency is
recorded beyond certain value of n, while in some other cases, it either decreases or
increases. Change in the stiffness and frequency mode is the probable reasons for this type
of tendency.
227
FGSSP having homogeneous core for different volume fraction index and skew angle values
(Type 1)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.4803 1.0897 1.4124 1.0056 1.0056 1.0056 1.0056 1.0056
0.1 1.0056 1.0056 1.0056 1.0056 1.0056 1.0056 1.0056 1.0056
0.2 1.0056 1.0056 1.0056 1.0056 1.0056 1.0056 1.0056 1.0056
0° 0.01 1.0544 1.2492 1.0524 1.0524 1.0524 1.0524 1.0524 1.0524

0.1 1.0524 1.0524 1.0524 1.0524 1.0524 1.0524 1.0524 1.0524
0.2 1.0524 1.0524 1.0524 1.0524 1.0524 1.0524 1.0524 1.0524
45° 0.01 1.1114 1.1838 1.1369 1.0490 1.0490 1.0060 1.0060 1.0060
0.1 1.0060 1.0060 1.0060 1.0060 1.0060 1.0060 1.0060 1.0060
0.2 1.0060 1.0060 1.0060 1.0060 1.0060 1.0060 1.0060 1.0060
60° 0.01 1.1431 1.1431 1.3165 1.0089 1.9838 1.9838 1.9838 1.7455
0.1 1.7455 1.7455 1.7455 1.7455 1.7455 1.7455 1.7455 1.7455
0.2 1.9838 1.9838 1.9838 1.9838 1.9838 1.9838 1.9838 1.9838

(Type 2)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 2.0972 1.2752 1.5329 1.5329 1.8944 1.0822 1.1903 1.0680
0.1 1.0680 1.1903 1.1903 1.1903 1.1903 1.1903 1.1903 1.1903
0.2 1.1903 1.1903 1.1903 1.1903 1.1903 1.1903 1.1903 1.1903
30° 0.01 1.0045 1.9647 1.9446 1.9150 1.8676 1.8300 1.7829 1.7825
0.1 1.6431 1.6241 1.6083 1.5843 1.5442 1.5111 1.4678 1.4674
0.2 1.2335 1.2335 1.6329 1.2424 1.1460 1.0557 1.0557 1.1076
45° 0.01 1.1076 1.1076 1.1076 1.1076 1.1076 1.1076 1.1076 1.1076
0.1 1.1076 1.1076 1.1076 1.1076 1.1076 1.1076 1.1076 1.1076
0.2 1.2103 1.2103 1.2103 1.2103 1.2103 1.3549 1.1317 1.1317
60° 0.01 1.1317 1.1317 1.1317 1.1317 1.1317 1.1317 1.1317 1.1317
0.1 1.1317 1.1317 1.1317 1.1317 1.1317 1.1317 1.1317 1.1317
0.2 2.0972 1.2752 1.5329 1.5329 1.8944 1.0822 1.1903 1.0680
Common observations can be obtained for both Type A and Type B FGSSP from the tables
provided in this sub section. For both the types of sandwich skew plates, as the thickness ratio
increases frequency decreases irrespective of the value of skew angle and volume fraction.
Similarly, if the value of volume fraction index increases no definite trend is observed in frequency
values. This is due to the change in stiffness of constituents at each layer.
228
(Type 3)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.2609 1.6644 1.0228 1.1175 1.0594 1.0314 1.0863 1.0863
0.1 1.0863 1.0863 1.0863 1.0863 1.0863 1.0863 1.0863 1.0863
0.2 1.0863 1.0863 1.0863 1.0863 1.0863 1.0863 1.0863 1.0863
30° 0.01 1.5492 1.0808 1.3180 1.1415 1.0710 1.0649 1.0050 1.1193
0.1 1.1193 1.1193 1.1193 1.1193 1.1193 1.1193 1.1193 1.1193
0.2 1.1193 1.1193 1.1193 1.1193 1.1193 1.1193 1.1193 1.1193
45° 0.01 1.4173 1.1003 1.1384 1.0268 1.0268 1.0679 1.0679 1.0679
0.1 1.0679 1.0679 1.0679 1.0679 1.0679 1.0679 1.0679 1.0679
0.2 1.0679 1.0679 1.0679 1.0679 1.0679 1.0679 1.0679 1.0679
60° 0.01 1.4071 1.2921 1.2019 1.1566 1.1566 1.2571 1.2571 1.2772
0.1 1.2772 1.2772 1.2772 1.2772 1.2772 1.2772 1.2772 1.2772
0.2 1.2772 1.2772 1.2772 1.2772 1.2772 1.2772 1.2772 1.2772

(Type 4)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.0753 1.0753 1.0554 1.0554 1.0076 1.0076 1.0076 1.0069
0.1 1.0069 1.0069 1.0069 1.0069 1.0069 1.0069 1.0069 1.0069
0.2 1.0069 1.0069 1.0069 1.0069 1.0069 1.0069 1.0069 1.0069
30° 0.01 1.2448 1.0868 1.1200 1.1581 1.0159 1.0159 1.0159 1.0159
0.1 1.0159 1.0159 1.0159 1.0159 1.0159 1.0159 1.0159 1.0159
0.2 1.0159 1.0159 1.0159 1.0159 1.0159 1.0159 1.0159 1.0159
45° 0.01 1.1796 1.0801 1.0752 1.0941 1.0147 1.0147 1.0147 1.0147
0.1 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147
0.2 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147 1.0147
60° 0.01 1.0011 1.0105 1.0105 1.0105 1.0105 1.0105 1.0036 1.0036
0.1 1.0036 1.0036 1.0036 1.0036 1.0036 1.0036 1.0036 1.0036
0.2 1.0036 1.0036 1.0036 1.0036 1.0036 1.0036 1.0036 1.0036
4.7.2 Free vibration frequency of FGSSS by considering different

boundary conditions, thickness schemes and skew angles
In this section, free vibration results are generated for FGSSS having different thickness
schemes. For this purpose, four kinds of sandwich shells that represent the influence of
core thickness with respect to overall thickness of the shell are assumed. In addition,
boundary conditions, skew angles, and thickness of FGSSS are varied to sketch their
influences on frequency parameter.
229
4.7.2.1 Influence of boundary conditions on frequency parameter of
FGSSS
Table 4.73: Influence of boundary conditions on frequency values of FGSSS having graded
core ( =15°)
Types Boundary n
condition Ceramic 2 5 10 Metal
Type 1 SSSS 4.9154 3.9675 3.5999 3.4426 3.2867
CCCC 3.2867 3.2867 4.2633 4.0876 3.9066
SCSC 4.1150 3.5737 3.3037 3.1728 3.0358
CFCF 1.8923 1.3716 1.1992 1.1355 1.0775
Type 2 SSSS 4.9054 3.7162 3.3125 3.1494 2.9916

CCCC 5.4645 4.3564 3.9187 3.7160 3.5059
SCSC 4.1063 3.3381 3.0142 2.8581 2.6933
CFCF 1.9012 1.2925 1.1250 1.0747 1.0356
Type 3 SSSS 4.8782 4.2694 3.9572 3.8123 3.6637

CCCC 3.6637 4.9740 4.6751 4.5294 4.3776
SCSC 4.1160 3.8544 3.6451 3.5401 3.4295
CFCF 1.8609 1.4670 1.3023 1.2295 1.1556
Type 4 SSSS 4.6328 4.7328 4.5780 4.4966 4.4112

CCCC 4.4112 4.4112 5.3608 5.2872 5.2101
SCSC 4.0703 4.3265 4.2344 4.1832 4.1287
CFCF 1.6973 1.5992 1.5021 1.4521 1.3991
core ( =30°)
Types Boundary n
Type 1 SSSS 5.0700 4.3841 3.9942 3.8220 3.6498
CCCC 3.6498 3.6498 4.5244 4.3333 4.1389
SCSC 4.9133 4.2096 3.8777 3.7205 3.5576
CFCF 1.9502 1.4347 1.2583 1.1921 1.1313
Type 2 SSSS 5.0269 4.0981 3.6692 3.4873 3.3059

CCCC 5.9108 4.6442 4.1655 3.9526 3.7361
SCSC 4.9029 3.9327 3.5428 3.3610 3.1721
CFCF 1.9593 1.3530 1.1802 1.1268 1.0841
Type 3 SSSS 5.1307 4.7244 4.3911 4.2329 4.0688

CCCC 4.0688 5.2990 4.9570 4.7928 4.6225
SCSC 4.9114 4.5402 4.2759 4.1451 4.0074
CFCF 1.9188 1.5324 1.3650 1.2903 1.2143
Type 4 SSSS 5.0897 5.2530 5.0850 4.9955 4.9008

CCCC 4.9008 5.8463 5.6839 5.5973 5.5064
SCSC 4.8089 5.0740 4.9507 4.8830 4.8111
CFCF 1.7618 1.6717 1.5743 1.5239 1.4703
Influence of boundary conditions on non-dimensional frequency values of FGSSS having

graded and homogeneous core is studied in this sub section and corresponding results are
presented in Tables 4.73-4.76 and Tables 4.77-4.80, respectively.
230
core ( =45°)
Types Boundary n
Type 1 SSSS 5.6051 4.8812 4.5167 4.3405 4.1565
CCCC 4.1565 4.1565 5.2527 5.0194 4.7870
SCSC 6.0783 5.0975 4.6669 4.4713 4.2723
CFCF 2.0161 1.5267 1.3478 1.2787 1.2143
Type 2 SSSS 5.5535 4.5174 4.0840 3.8770 3.6598

CCCC 7.0858 5.4430 4.8540 4.6116 4.3752
SCSC 6.0658 4.7590 4.2686 4.0540 3.8378
CFCF 2.0245 1.4403 1.2621 1.2038 1.1546
Type 3 SSSS 5.6802 5.3364 5.0414 4.8909 4.7282

CCCC 4.7282 4.7282 4.7282 4.7282 5.3050
SCSC 6.0739 5.5097 5.1557 4.9846 4.8053
CFCF 1.9869 1.6287 1.4617 1.3860 1.3083
Type 4 SSSS 5.7431 6.0897 5.9351 5.8477 5.7524

CCCC 5.7524 5.7524 5.7524 5.7524 5.7524
SCSC 5.7524 6.1876 6.0144 5.9208 5.8214
CFCF 1.8509 1.7816 1.6863 1.6366 1.5835
core ( =60°)
Types Boundary n
Condition Ceramic 2 5 10 Metal
Type 1 SSSS 7.5859 6.3890 5.8599 5.6177 5.3689
CCCC 10.3703 8.4203 7.5711 7.2115 6.8598
SCSC 8.9551 7.3417 6.6552 6.3579 6.0624
CFCF 1.9711 1.5987 1.4344 1.3657 1.2992
Type 2 SSSS 7.5309 5.9078 5.3004 5.0382 4.7763

CCCC 10.4025 7.9369 7.0353 6.6915 6.3719
SCSC 8.9465 6.8604 6.1095 5.8103 5.5247
CFCF 1.9768 1.5080 1.3363 1.2713 1.2105
Type 3 SSSS 7.6533 7.0008 6.5715 6.3602 6.1337

CCCC 6.1337 6.1337 6.1337 6.1337 6.1337
SCSC 6.1337 6.1337 7.3481 7.0704 6.7787
CFCF 1.9521 1.7043 1.5575 1.4876 1.4145
Type 4 SSSS 7.6808 8.0897 7.8792 7.7638 7.6394

CCCC 7.6394 7.6394 7.6394 7.6394 7.6394
SCSC 7.6394 7.6394 8.6486 8.4922 8.3247
CFCF 1.8883 1.8838 1.8042 1.7610 1.7141
To perform the numerical analysis, skew angles from 15° to 60° and four boundary
conditions (simply supported, clamped, simply supported-clamed, clamped-free) are
considered. As the skew angle increases, the frequency value increases for all kinds of
FGSSS considered. In all the cases it is observed that increase in skew angle raises the
value of frequency parameter and this tendency is irrespective of the thickness schemes
and values of volume fraction index.
231
Table 4.77: Influence of boundary conditions on frequency values of FGSSS having
Types Boundary n
Type 1 SSSS 3.8651 3.4771 3.2365 3.0474 2.8077
CCCC 4.5201 4.1951 3.9970 3.8436 3.6475
SCSC 3.2985 3.0225 2.8497 2.7124 2.5343
CFCF 1.4602 1.4007 1.3490 1.2978 1.2227
Type 2 SSSS 3.5417 3.2889 3.1427 3.0350 2.9062

CCCC 2.9062 2.9062 2.9062 2.9062 2.9062
SCSC 2.9062 2.9062 2.9062 2.9062 2.9062
CFCF 2.9062 2.9062 2.9062 2.9062 2.9062
Type 3 SSSS 4.2553 3.7135 3.3632 3.0738 2.6822

CCCC 2.6822 2.6822 2.6822 2.6822 2.6822
SCSC 2.6822 2.6822 2.6822 2.6822 2.6822
CFCF 2.6822 2.6822 2.6822 2.6822 2.6822
Type 4 SSSS 4.7222 4.0225 3.5341 3.1454 2.5445

CCCC 2.5445 2.5445 2.5445 2.5445 2.5445
SCSC 2.5445 2.5445 2.5445 2.5445 2.5445
CFCF 2.5445 2.5445 2.5445 2.5445 2.5445

Types Boundary n
Type 1 SSSS 3.9255 3.5277 3.2814 3.0877 2.8418
CCCC 2.8418 2.8418 2.8418 2.8418 2.8418
SCSC 2.8418 2.8418 2.8418 2.8418 2.8418
CFCF 2.8418 2.8418 2.8418 2.8418 2.8418
Type 2 SSSS 3.6094 3.3539 3.2054 3.0957 2.9642

CCCC 2.9642 2.9642 2.9642 2.9642 2.9642
SCSC 2.9642 2.9642 2.9642 2.9642 2.9642
CFCF 2.9642 2.9642 2.9642 2.9642 2.9642
Type 3 SSSS 4.3300 3.7803 3.4251 3.1330 2.7364

CCCC 2.7364 2.7364 2.7364 2.7364 2.7364
SCSC 2.7364 2.7364 2.7364 2.7364 2.7364
CFCF 2.7364 2.7364 2.7364 2.7364 2.7364
Type 4 SSSS 4.9692 4.1445 3.6149 3.2118 2.6032

CCCC 2.6032 2.6032 2.6032 2.6032 2.6032
SCSC 2.6032 2.6032 2.6032 2.6032 2.6032
CFCF 2.6032 2.6032 2.6032 2.6032 2.6032
Among the different types of sandwich schemes considered, when core is modeled with
FGM material, Type 4 shells ensures highest frequency values in most of the cases. While
in all the cases, Type 4 shells record maximum frequency value if the core is replaced with
homogeneous part. Effect of volume fraction index is to decrease the stiffness of the shell
to lower the frequency value and hence results in lower value of frequency.
232
Types Boundary n
Type 1 SSSS 4.3375 3.9071 3.6374 3.4243 3.1515
CCCC 3.1515 3.1515 3.1515 3.1515 3.1515
SCSC 3.1515 3.1515 3.1515 3.1515 3.1515
CFCF 3.1515 3.1515 3.1515 3.1515 3.1515
Type 2 SSSS 3.9936 3.7168 3.5539 3.4327 3.2865

CCCC 3.2865 3.2865 3.2865 3.2865 3.2865
SCSC 3.2865 3.2865 3.2865 3.2865 3.2865
CFCF 3.2865 3.2865 3.2865 3.2865 3.2865
Type 3 SSSS 4.7623 4.1619 3.7745 3.4608 3.0309

CCCC 3.0309 3.0309 3.0309 3.0309 3.0309
SCSC 3.0309 3.0309 3.0309 3.0309 3.0309
CFCF 3.0309 3.0309 3.0309 3.0309 3.0309
Type 4 SSSS 5.4658 4.5393 3.9412 3.5132 2.8804

CCCC 2.8804 2.8804 2.8804 2.8804 2.8804
SCSC 2.8804 2.8804 2.8804 2.8804 2.8804
CFCF 2.8804 2.8804 2.8804 2.8804 2.8804

Types Boundary n
Type 1 SSSS 5.8678 5.3140 4.9641 4.6887 4.3319
CCCC 8.8720 8.2168 7.7848 7.4368 6.9698
SCSC 8.8720 8.2168 7.7848 7.4368 6.9698
CFCF 8.8720 8.2168 7.7848 7.4368 6.9698
Type 2 SSSS 5.4450 5.0910 4.8794 4.7207 4.5277

CCCC 4.5277 4.5277 4.5277 4.5277 7.2658
SCSC 8.4081 7.9831 7.7195 7.5172 7.2658
CFCF 8.4081 7.9831 7.7195 7.5172 7.2658
Type 3 SSSS 6.3409 5.5443 5.0421 4.6559 4.1225

CCCC 4.1225 4.1225 4.1225 4.1225 4.1225
SCSC 4.1225 4.1225 4.1225 4.1225 4.1225
CFCF 4.1225 4.1225 4.1225 4.1225 4.1225
Type 4 SSSS 7.1461 5.8461 5.0258 4.5269 3.8222

CCCC 3.8222 3.8222 3.8222 3.8222 3.8222
SCSC 3.8222 3.8222 3.8222 3.8222 3.8222
CFCF 3.8222 3.8222 3.8222 3.8222 3.8222
233
4.7.2.2 Influence of thickness-side ratio on frequency values of FGSSS
FGSSS having graded core for different volume fraction index and skew angle values(Type 1)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.3175 1.3175 1.3175 1.3175 1.1815 1.1815 1.1815 1.1535
0.1 1.1535 1.1535 1.1535 1.1815 1.1815 1.1815 1.1815 1.1815
0.2 1.1815 1.1815 1.1815 1.1815 1.1815 1.1815 1.1815 1.1815
30° 0.01 1.0486 1.0486 1.2150 1.1014 1.1014 1.1148 1.1185 1.1546
0.1 1.1546 1.1546 1.1546 1.1546 1.1546 1.1546 1.1546 1.1546
0.2 1.1546 1.1546 1.1546 1.1546 1.1546 1.1546 1.1546 1.1546
45° 0.01 1.1014 1.1014 1.1014 1.1014 1.1014 1.1014 1.1014 1.0080
0.1 1.0080 1.0080 1.0080 1.0080 1.0080 1.0080 1.0080 1.0080
0.2 1.0080 1.0080 1.0080 1.0080 1.0080 1.0080 1.0080 1.0080
60° 0.01 1.1614 1.5136 1.6298 1.9067 1.1257 1.6212 1.6212 1.2225
0.1 1.2225 1.2225 1.2225 1.2225 1.2225 1.2225 1.2225 1.2225
0.2 1.2225 1.2225 1.2225 1.2225 1.2225 1.2225 1.2225 1.2225

FGSSS having graded core for different volume fraction index and skew angle values(Type 2)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.6936 1.6936 1.1288 1.0511 1.5198 1.5198 1.5198 1.4648
0.1 1.4648 1.4648 1.4648 1.4648 1.4648 1.4648 1.4648 1.4648
0.2 2.7165 2.7165 2.0562 2.0562 1.4753 1.4753 1.0029 1.0028
30° 0.01 1.0028 1.0028 1.0028 1.0028 1.0028 1.0028 1.0028 1.0028
0.1 1.0028 1.0028 1.0028 1.0028 1.0028 1.0028 1.0028 1.0028
0.2 1.0372 1.0372 1.0372 1.0372 1.0372 1.0372 1.0372 1.0372
45° 0.01 1.0372 1.0372 1.0372 1.0372 1.0372 1.0372 1.0372 1.0372
0.1 1.0372 1.0372 1.0372 1.0372 1.0372 1.0372 1.0372 1.0372
0.2 1.7139 1.2189 1.6974 1.5231 1.1085 1.1082 1.1082 1.1082
60° 0.01 1.1082 1.1082 1.1082 1.1082 1.1082 1.1082 1.1082 1.1082
0.1 1.1082 1.1082 1.1082 1.1082 1.1082 1.1082 1.1082 1.1082
0.2 1.6936 1.6936 1.1288 1.0511 1.5198 1.5198 1.5198 1.4648
Effect of thickness ratio (h/b) on fundamental frequency of FGSSS having graded and
homogeneous core is shown in Tables 4.81-4.84 and Tables 4.85-4.88, respectively, for
different kinds of sandwich schemes and skew angles.
234
FGSSS having graded core for different volume fraction index and skew angle values (Type
3)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.1879 1.1879 1.1879 1.0455 1.1089 1.1089 1.0443 1.0808
0.1 1.0808 1.0808 1.0808 1.0808 1.0808 1.0808 1.0808 1.0808
0.2 1.0808 1.0808 1.0808 1.0808 1.0808 1.0808 1.0808 1.0808
30° 0.01 1.1314 1.1314 1.1314 1.1314 1.1314 1.1314 1.0165 1.0165
0.1 1.0165 1.0165 1.0165 1.0165 1.0165 1.0165 1.0165 1.0165
0.2 1.0165 1.0165 1.0165 1.0165 1.0165 1.0165 1.0165 1.0165
45° 0.01 1.8822 1.0752 1.1848 1.1848 1.1848 1.1848 1.1286 1.0070
0.1 1.0070 1.0070 1.0070 1.0070 1.0070 1.0070 1.0070 1.0070

0.2 1.0070 1.0070 1.0070 1.0070 1.0070 1.0070 1.0070 1.0070
60° 0.01 1.1449 1.6214 1.6214 2.0521 1.0502 1.0502 1.0502 1.0567
0.1 1.0567 1.0567 1.0567 1.0567 1.0567 1.0567 1.0567 1.0567
0.2 1.0567 1.0567 1.0567 1.0567 1.0567 1.0567 1.0567 1.0567

FGSSS having graded core for different volume fraction index and skew angle values (Type
4)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.0967 1.0717 1.0119 1.0152 1.0152 1.0152 1.0152 1.0152
0.1 1.0152 1.0152 1.0152 1.0152 1.0152 1.0152 1.0152 1.0152
0.2 1.0152 1.0152 1.0152 1.0152 1.0152 1.0152 1.0152 1.0152
30° 0.01 1.2155 1.1256 1.0058 1.0205 1.0205 1.0205 1.0205 1.0782
0.1 1.0782 1.0782 1.0782 1.0782 1.0782 1.0782 1.0782 1.0782
0.2 1.0782 1.0782 1.0782 1.0782 1.0782 1.0782 1.0782 1.0782
45° 0.01 1.3451 1.3833 1.4253 1.1502 1.0470 1.0470 1.0470 1.0470
0.1 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470
0.2 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470 1.0470
60° 0.01 1.3956 1.1091 1.1039 1.0572 1.0202 1.0575 1.0913 1.0913
0.1 1.0913 1.0913 1.0913 1.0913 1.0913 1.0913 1.0913 1.0913
0.2 1.0913 1.0913 1.0913 1.0913 1.0913 1.0913 1.0913 1.0913
Frequency values correspond to thickness-side ratio (h/a) 0.1 and 0.2, shows equal free
vibration frequency values for all the skew angles and sandwich schemes. But for thin shell
cases, quite different results are observed for all the types of sandwich shells and no
definite frequency trend was noticed. Similarly, no stable response is observed in view of
the effect of volume fraction index on frequency. In some cases, as the volume fraction
index increases, the frequency values are increasing or decreasing depending upon the
stiffness effect. In general, for most of the cases, increase in skew angle increases the
235
frequency parameter and this tendency changes depending upon the values of thickness
and skew angles.
FGSSS having homogeneous core for different volume fraction index and skew angle values
(Type 1)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.2066 1.2422 1.2422 1.2422 1.2422 1.2422 1.2422 1.2422
0.1 1.2422 1.2422 1.2422 1.2422 1.2422 1.2422 1.2422 1.2422
0.2 1.2422 1.2422 1.2422 1.2422 1.2422 1.2422 1.2422 1.2422
30° 0.01 1.1149 1.1149 1.1107 1.1107 1.1107 1.0777 1.1107 1.1107
0.1 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107
0.2 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107 1.1107
45° 0.01 1.0919 1.2656 1.2656 1.2656 1.2656 1.0410 1.0410 1.0410
0.1 1.0410 1.0410 1.0410 1.0410 1.0410 1.0410 1.0410 1.0410
0.2 1.0410 1.0410 1.0410 1.0410 1.0410 1.0410 1.0410 1.0410
60° 0.01 2.3273 3.5123 1.0290 1.1460 1.1460 1.1460 1.1116 1.1116
0.1 1.1116 1.1116 1.1116 1.1116 1.1116 1.1116 1.1116 1.1116
0.2 1.1116 1.1116 1.1116 1.1116 1.1116 1.1116 1.1116 1.1116

(Type 2)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.1173 1.5029 1.5134 1.3894 1.0298 1.0298 1.5134 1.5134
0.1 1.5134 1.5134 1.5134 1.5134 1.5134 1.5134 1.5134 1.5134
0.2 1.5134 1.5134 1.5134 1.5134 1.5134 1.5134 1.5134 1.5134
30° 0.01 1.1535 1.1535 1.1535 1.1535 1.1535 1.1535 1.1535 1.1535
0.1 1.1535 1.1535 1.1535 1.1535 1.1535 1.1535 1.1535 1.1535
0.2 1.2039 1.0304 1.0304 1.0039 1.0304 1.0304 1.0304 1.0304
45° 0.01 1.0304 1.0304 1.0304 1.0304 1.0304 1.0304 1.0304 1.0304
0.1 1.4147 1.0049 1.0049 1.0049 1.0049 1.5502 1.0001 1.0001
0.2 1.0001 1.0001 1.0001 1.0001 1.0001 1.0001 1.0001 1.0001
60° 0.01 1.0001 1.0001 1.0001 1.0001 1.0001 1.0001 1.0001 1.0001
0.1 1.1173 1.5029 1.5134 1.3894 1.0298 1.0298 1.5134 1.5134
0.2 1.5134 1.5134 1.5134 1.5134 1.5134 1.5134 1.5134 1.5134

(Type 3)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.2999 1.1366 1.3245 1.2390 1.3593 1.1411 1.1144 1.1144
0.1 1.1144 1.1144 1.1144 1.1144 1.1144 1.1144 1.1144 1.1144
0.2 1.1144 1.1144 1.1144 1.1144 1.1144 1.1144 1.1144 1.1144
30° 0.01 1.2950 1.6649 1.0884 1.0884 1.0007 1.1304 1.0155 1.0155
0.1 1.0155 1.0155 1.0155 1.0155 1.0155 1.0155 1.0155 1.0155
0.2 1.0155 1.0155 1.0155 1.0155 1.0155 1.0155 1.0155 1.0155
45° 0.01 1.6666 1.2480 1.0874 1.0874 1.0565 1.3196 1.1989 1.1589
0.1 1.1589 1.1589 1.1589 1.1589 1.1589 1.1589 1.1589 1.1589
0.2 1.1589 1.1589 1.1589 1.1589 1.1589 1.1589 1.1589 1.1589
60° 0.01 2.0178 2.4068 1.2956 1.2956 1.3892 1.0787 1.0787 1.2950
0.1 1.2950 1.2950 1.2950 1.2950 1.2950 1.2950 1.2950 1.2950
0.2 1.2950 1.2950 1.2950 1.2950 1.2950 1.2950 1.2950 1.2950
236
(Type 4)
n
 h/b
0 0.5 1 2 5 10 50 1000
15° 0.01 1.3076 1.2094 1.1310 1.0655 1.0655 1.0655 1.0602 1.0602
0.1 1.0602 1.0602 1.0602 1.0602 1.0602 1.0602 1.0602 1.0602
0.2 1.0602 1.0602 1.0602 1.0602 1.0602 1.0602 1.0602 1.0602
30° 0.01 1.3774 1.0136 1.2687 1.0359 1.0359 1.0359 1.0359 1.0359
0.1 1.0359 1.0359 1.0359 1.0359 1.0359 1.0359 1.0359 1.0359
0.2 1.0359 1.0359 1.0359 1.0359 1.0359 1.0359 1.0359 1.0359
45° 0.01 1.1063 1.1063 1.1063 1.0925 1.0244 1.0244 1.0244 1.0205
0.1 1.0205 1.0205 1.0205 1.0205 1.0205 1.0205 1.0205 1.0205
0.2 1.0205 1.0205 1.0205 1.0205 1.0205 1.0205 1.0205 1.0205
60° 0.01 1.3532 1.3565 1.4050 1.3628 1.1206 1.0997 1.1118 1.1118
0.1 1.1118 1.1118 1.1118 1.1118 1.1118 1.1118 1.1118 1.1118
0.2 1.1118 1.1118 1.1118 1.1118 1.1118 1.1118 1.1118 1.1118
4.8 GEOMETRICALLY NON-LINEAR (GNL) ANALYSIS OF FGSP

AND FGSS
4.8.1 GNL of FGSP by considering different thickness schemes
Under this section, several numerical examples are solved to demonstrate the non-linear
responses of FGSP having different thickness schemes. Two set of combinations of
ceramic and metal constituents are considered for FGSP. FGSP having graded core
consists of Al and ZrO2 as constituents, and Ti-6Al-4V and Al2O3 are taken as constituent
materials for FGSP having graded face sheets. A mesh division of 12x12 is found to be
sufficient for the convergence of the results and hence implemented in all the subsequent
examples of GNL analysis. A FGSP with simply supported boundary at all the edges is
considered in the study. However, non-linear finite element code developed does not pose
any limitations on other type of boundary conditions. Since no data is available in the
literature to validate the non-linear response of FGSP, the present FE formulation has
been compared with ordinary isotropic and functionally graded plates (FGP) and then
new results related to FGSP incorporating FGM as core/skin are presented. The following
non-dimensional parameters are used in GNL analysis.
w
Deflection under uniformly distributed loading: w 
h
q
Loading parameter: q0 
Em h 4
z
Thickness co-ordinate: z 
h
 xx
Axial stresses:  xx 
q
237
4.8.1.1 Validation of results for geometrically non-linear analysis of
isotropic and FGM plates subjected to mechanical loading
In the first validation study, an isotropic square plate subjected to a uniform pressure of
intensity q N/m2 with clamped boundary all along the edges is considered. The geometric
and material properties of the plate are: a = b = 300 in., h = 3in., E = 0.3x108 lb/in2, 𝛾 =
0.316. Due to the symmetry, only one quarter of the plate is analyzed. Figure 4.48
demonstrates the variation of non-dimensional central deflection with load parameter q
for isotropic case. The present HSDT results were compared with the results of Pica et al.
(1979) based on Mindlin theory (FSDT) and close agreement between the results is
observed. A loading range of 17.79 N/m2 to 402 N/m2 is applied on the top of the plate to
trace the complete load-deflection equilibrium path.
As the second validation example, the Aluminium/zirconia FG plate is considered
to validate the present HSDT results. The FG plate under consideration have a side length
of a = b = 0.2 m and thickness h = 0.01 m. From Figure 4.49, it is manifested that the
present results compares very well with the data reported by Praveen and Reddy (1998).
Both the studies incorporate TSDT in conjunction with the finite element formulation
taking von Kármán assumptions in the Green strain relation. Results were plotted for
both isotropic and graded plates. For graded plates the value of volume fraction index
chosen are: n= 0.2, 0.5, 1.0 and 2.0. It is observed that, except for the isotropic case (pure
ceramic plate), the results for other cases compares very well with the present cases.
2.0
1.6
1.2
Pica et al. (1979)

Present
0.8
0.4
0 100 200 300 400
Non-dimensinal load parameter (P)
Figure 4.48 Validation of non-linear load-displacement response of isotropic plates
238
n=1000 (Praveen and Reddy 1998), n=1000 (present),
n=0 (Praveen and Reddy 1998), n=0 (present),
n=0.2 (Praveen and Reddy 1998), n=0.2 (present),
0.22
n=2.0 (Praveen and Reddy 1998), n=2.0 (present)
0.20
N0n-dimensional displacement (w)

0.18
0.16
0.14
0.12
0.10
0.08
0.06
0.04
0.02
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Non-dimensinal load parameter (P)
Figure 4.49 Validation of non-linear load-displacement response of FGM plates
4.8.1.2 Non-linear load-displacement response of FGSP having graded

core subjected to mechanical loading
0.20
n=0 0.24 n=0
0.18 n=0.5 n=0.5
n=1 0.22
n=1
Non-dimensional displacment (w)
0.16 n=5 n=5

0.20
n=10 n=10
0.14 n=1000 0.18 n=1000
0.12 0.16
0.14
0.10
0.12
0.08
0.10
0.06
0.08
0.04
0.06
0.02 0.04
1 2 3 4 5 1 2 3 4 5
Non-dimensional load parameter (P) 0.22 Non-dimensional load parameter (P)
0.30 n=0 (a) Type A FGSP n=0 (b) Type B FGSP
n=0.5 0.20 n=0.5
0.27 n=1 n=1
n=5 0.18
n=5
0.24 n=10 n=10
0.16
n=1000 n=1000
0.21
0.14
0.18
0.12
0.15
0.10
0.12
0.08
0.09
0.06
0.06
0.04
0.03
0.02
1 2 3 4 5 1 2 3 4 5
Non-dimensional load parameter (P) Non-dimensional load parameter (P)
(c) Type C FGSP (d) Type D FGSP
Figure 4.50 Non-linear load-displacement response of FGSP having graded core by

considering various thickness schemes and volume fraction indexes (n) subjected to loading
magnitude of 1-5 N/mm2
239
5.0
n=0 n=0
4.5
4.5 n=0.5 n=0.5
n=1 n=1

n=5 4.0
n=5
4.0
n=10 n=10
n=1000 n=1000
3.5
3.5
3.0
3.0
2.5
2.5
2.0
2.0
1.5
1.0 1.5
0 200 400 600 800 1000 0 200 400 600 800 1000
n=0 (a) Type A FGSP (b) Type B FGSP
4.5 n=0
n=0.5 4.5
n=0.5
n=1 n=1
4.0 n=5 4.0 n=5

n=10 n=10
n=1000 n=1000
3.5 3.5
3.0
3.0
2.5
2.5
2.0
2.0
1.5
1.5
0 200 400 600 800 1000
0 200 400 600 800 1000
Non-dimensional load parameter (P)
(d) Type D FGSP
(c) Type C FGSP

considering various thickness schemes and volume fraction indexes (n) subjected to load
Non-linear load-deflection path for FGSP having core as FGM and subjected to
uniformly distributed load is depicted in Figure.4.50. Four classes of thickness schemes
viz., 1-8-1, 1-1-1, 2-1-2 and 1-2-1 are chosen for the study and the schemes are
represented as Type A, Type B, Type C and Type D FGSP, respectively. Same notations
are used in all the results presented under this sub section. Note that the thickness of each
layer is designated from the bottom of the plate. Results were shown for pure ceramic and
metal plates in addition to graded plates with the value of n = 0.5, 1.0, 5.0 and 10.0. A
significant deviation between isotropic and graded plates is recorded for Type A FGSP
under different values of volume fraction indexes. This tendency is found to be quite
240
opposite while thickness scheme 2-1-2 is chosen for the study. The thickness ratio of
ceramic and metal plates i.e., pure isotropic plates is considerably higher than the core
thickness (two times the thickness of core portion). Thus the role of volume fraction
index in defining the gradient property of the plate becomes least significant in this case.
Similar reason can be stated for Type A FGSP for exhibiting higher deviation among the
different cases.
Further it is observed that the plates identified by graded property i.e., core
portion have shown intermediate response between those of pure ceramic and pure metal
plates in all the cases of thickness schemes. In Figure 4.51, the load-deflection path is
traced for Al/ZrO2 plate with identical material and geometric properties as in Figure
4.50, except that the magnitude of load parameter is higher in this case. A loading range
of 100-1000 N/m2 was selected to observe the non-linear response in FGSP. Considerable
deviation is noticed between Figure 4.50 and Figure 4.51 for isotropic and graded plates.
For instance as far as Type C plate is considered in Figure 4.51, the difference in non-
linear responses of the isotropic and graded plates are more visible than the
corresponding one shown in Figure 4.50. Thus it is interpreted that the loading
pattern/magnitude imposed on the plate has important role on non-linear response of
FGSP.
4.8.1.3 Influence of thickness ratio (a/h) on non-linear load-displacement

response of FGSP having graded core subjected to mechanical loading
Influence of thickness ratio (a/h) on non-linear response of the FGSP is demonstrated in
Figure 4.52. To show the effect of gradation parameter two types of volume fraction
index n= 1.0 and 2.0 are chosen. When thickness ratio (a/h) assumes the value around
30.0, a sharp decline in the deflection value is noticed, beyond which a stable behavior is
noticed for all the thickness schemes under consideration. Pure metallic plates exhibit
high deflection values compared to pure ceramic and graded plates due to their low
stiffness values. Again, an intermediate response of graded plates is observed similar to
previous cases. Among all the types of FGSP, higher value of deflection parameter is
noticed for Type A FGSP.
241
4.8
n=0
4.5 n=2
4.6 n=1000

n=1
n=0
4.4
n=2
4.4
n=1000
n=1
4.3
4.2
4.2
4.0
4.1
3.8
4.0
0 40 80 120 160 0 40 80 120 160
(b) Type B FGSP
(a) Type A FGSP n=0
n=0 4.6
4.45
n=2
n=2 n=1000
n=1000 4.5 n=1
4.40 n=1
4.4
4.35
4.3
4.30 4.2
4.1
4.25
4.0
4.20
3.9
4.15
0 40 80 120 160
0 40 80 120 160
Figure 4.52 Influence of thickness ratio (a/h) ratio on non-linear bending response of FGSP
having graded layer as core
4.8.1.4 Non-linear load-displacement response of FGSP having

homogeneous core subjected to mechanical loading
To perform non-linear analysis for FGSP having graded face sheet, Ti-6Al-4V/Aluminium
oxide is considered as ceramic/metal constituents of the plate. The other geometric
properties of the plate are same as those taken for FGSP having FG core. A loading
parameter within the range of 1-5 N/m2 is chosen to represent the load-deflection path. As
observed in previous cases i.e., for FGSP with FG core, linear response is noticed for all
the thickness schemes. Different values of volume fraction index are selected so as to
represent both isotropic and graded plates in the plate geometry.
242
0.10
0.14
0.09

0.12
0.08
0.07 0.10
0.06
0.08 n=0
0.05 n=0.5
0.06 n=1
0.04 n=5
n=0 n=10
0.03 0.04
n=1
n=5
0.02 n=10 0.02
0.01
0.00
1 2 3 4 5 1 2 3 4 5
(a) Type A FGSP (b) Type B FGSP
0.20
n=0
0.18 n=0.5 0.10 n=0
n=1 n=0.5
0.16 n=5 0.09

n=1
n=10 Non-dimensional displacement (w) 0.08
n=5
0.14
0.07
0.12
0.06
0.10
0.05
0.08
0.04
0.06
0.03
0.04
0.02
0.02 0.01
0.00 0.00
1 2 3 4 5 1 2 3 4 5
Figure 4.53 Non-linear load-displacement response of FGSP having homogeneous core by
As the n value increases the ceramic content increases i.e., when n =5.0 and 10.0, marked
response of the plate from that of isotropic plates is observed. In particular this type of
behavior is more obvious in Type B and Type D FGSP (Figure.4.53). For Type C plates,
where the thickness of the skin layers are twice that of thickness of core and is defined by
the pure ceramic content, considerable deviation is observed between isotropic and graded
plates. On contrary to this statement, 1-2-1 thickness scheme (Type D plate) shows least
difference in the responses among isotropic and graded plates. This tendency in the
behavior of FGSPs incorporating various thickness schemes clearly indicates the influence
of thickness parameter on non-linear response of FGSP.
243
Of various thickness schemes chosen for the non-linear analysis, Type C plates
records maximum deflection under the action of uniform pressure load. Further in Type B
and Type D FGSP, value of power law exponent, n=10 gives insignificant deflection due
to the high stiffening effect of the plate. To get the non-linear response of the FGSP,
loading with higher magnitude is chosen for the analysis and the corresponding load-
displacement path is depicted in Figure.4.54. It may be observed that for 1-1-1 thickness
scheme, for lower values of volume fraction indexes (i.e., n= 0, 0.5 and 1.0) will not play
any significant role in dictating the response of the plate, and this tendency is quite
opposite in all the other cases of thickness schemes (Type A, Type C and Type D). Also in
Type D FGSP, rise in the value of power law exponent tends to give more deflection which
is not evident in Type B and Type C FGSPs.
3.5 3.5
n=0
3.0 n=0.5
3.0
n=1
n=5
2.5 n=10
2.5
2.0
2.0
1.5
n=0
n=1
n=5 1.0
1.5
n=10
0.5
1.0
0.0
0 200 400 600 800 1000 0 200 400 600 800 1000

n=0 (a) Type A FGSP n=0 (b) Type B FGSP
3.5 4.5
n=0.5 n=0.5
n=1 n=1
3.0 n=5 4.0 n=5

n=10 n=10
2.5 3.5
2.0 3.0
1.5 2.5
1.0 2.0
0.5 1.5
0 200 400 600 800 1000 0 200 400 600 800 1000


244
4.8.1.5 Influence of thickness ratio (a/h) on non-linear load-displacement
response of FGSP having homogeneous core subjected to mechanical
loading
Influence of each chosen thickness ratio on non-linear response of the FGSP is investigated
in Figure 4.55. Up to the value of a/h=200, a sharp declining tendency is noticed and
further which an asymptotic response is visualized in the FGSPs. Except for Type D
FGSP, the plate with the linear value of volume fraction index (n=1.0) gives highest
deflection for all other FGSPs. Further a major role of volume fraction index on non-linear
deflection is observed for all the thickness schemes.
3.32
3.32
3.28
3.28
3.24
3.24
3.20 3.20
n=0
n=2 3.16 n=0
3.16
n=1 n=2
3.12 n=1
3.12
3.08
3.08
3.04
0 200 400 600 800 1000 0 200 400 600 800 1000
Thickness ratio (a/h) 3.3 Thickness ratio (a/h)

3.48 (a) Type A FGSP (b) Type B FGSP
3.2
3.44 3.1
3.0
3.40
2.9 n=0
n=2
2.8 n=1
3.36
n=0
n=2 2.7
n=1
3.32
2.6
0 200 400 600 800 1000 0 200 400 600 800 1000

(c) Type C FGSP (a) Type A FGSP
Figure 4.55 Influence of thickness ratio (a/h) on non-linear bending response of FGSP
having graded layer as core
245
4.8.2 Geometrically non-linear analysis of FGM cylindrical, spherical,
hyperbolic paraboloid, and hypar shells subjected to mechanical
loading
Geometrically non-linear analysis (GNL) of functionally graded shells is presented under
this sub section by considering various shell geometries. Limited numbers of results are
available for geometrically non-linear anlaysis of FGM shells and the available results are
based on FSDT model. Hence to accurately predict the transverse displacement response
of FGM shells, HSDT model is implemented in the present GNL analysis. Initially, the
validation is performed by comparing the present HSDT results with those of the
available results based on FSDT. Further, no results are available in the literature for shell
incorporating the twist radii of curvature. Subsequently, new results are generated for
different shell forms and reported in different figures. For validation, Al/ZrO2 shell is
considered and to generate new results, Ti-6V-Al2O3/Al shell is considered. The
geometric properties, a = b = 0.2 m and h= 0.01m are considered for all the shell forms.
Due to the symmetry of geometry, only quarter of the shell is analyzed.
4.8.2.1 Validation of results for Al/ZrO2 FGM cylindrical shell subjected

to mechanical loading
Zhao et al. (200b) (ceramic)

Present (ceramic)
Zhao et al. (200b) (metal)
1.2 Present (metal)
Zhao et al. (200b) (n=0.5)
1.0 Present (n=0.5)

Zhao et al. (200b) (n=2)
Present (n=2.0)
0.8
0.6
0.4
0.2
0.0
5 10 15 20 25 30
Figure 4.56 Non-linear load-deflection graph for GNL of FGM cylindrical shells subjected to
mechanical loading
246
For validation, non-linear bending analysis of simply supported Aluminium/Zirconia shell
considered by Zhao et al., (2009b) is studied. As depicted in Figure 4.56, the present
results based on HSDT show excellent correlation with the published results during the
initial loading conditions and however show deviation for higher magnitude of loading.
Since the non-linear phenomena is basically iteration based process, the quantity of error
seems to be negligible at initial steps and accumulates further during subsequent iterations.
Other probable reasons that may be responsible for the discrepancies between the results
may be the assumption of different kinematic models and the way of implementation of
curvature terms in corresponding strain part.
In reference paper (Zhao et al., 2009b), FSDT is used to define the FGM model in
terms of single radii of curvature. But in the present model, to capture the membrane-
bending coupling in FGM the flexural and deformation terms are incorporated in the
displacement model by incorporating all the three radii of curvature in the formulation. In
addition, the solution methodologies adopted by the two authors may be another cause
for divergence between the results. Since the realistic parabolic variation of transverse
displacement is assumed in the present study, the current results appear to be lower than
the reference results.
4.8.2.2 Geometrically non-linear analysis of FGM cylindrical shells

GNL responses of FGM cylindrical shells incorporating different boundary conditions
and loading range are depicted in Figure 4.56. For illustration purpose, simply supported
and clamped boundary conditions and loading ranges 5-60N/m2 and 100-600 N/m2 are
chosen. When higher magnitudes of loadings are applied in incremental forms, non-linear
pattern of the displacement curves are observed. Ceramic and metal shells, respectively,
shows lower and higher values of deflection parameters; whereas FGM shells record
intermediate deflections. If the value of volume fraction index is increased beyond the
linear variation of ceramic concentration i.e., n = 1.0, considerable reduction of stiffness
is observed. Choosing the value of n beyond 1.0 has no impact on deflection parameter of
FGM cylindrical shell.
In Figure 4.58, in order to demonstrate the influence of curvature effect on non-
dimensional deflection of FGM cylindrical shell, various values of curvature and volume
fraction indexes are chosen. It is to be noted that, here Rx is taken as value of 1/Rx. When
247
Rx/a=5, i.e., Rx=0.2 (higher value of radius) shows maximum deflection among the group
of different values of Rx. As the value of Rx reduces, lower deflection is observed. Up to
value of n=30, a steep increase in deflection parameter is observed and for higher values
asymptotic response is observed in Figure 4.58. Through-the-thickness variation of in-
plane axial stress component is shown for FGM cylindrical shell in Figure 4.59. Simply
supported and clamped boundary conditions are chosen for moderately thick (h=0.1 m)
and thin (h=0.001 m) shells. Tensile stresses are observed at the top of the shell, while
compressive stresses are observed at the bottom. Further, it is observed that the variation
of axial stress in thick shells is quite opposite as compared to thin shell case.
0.6 n=0 3.6 n=0

n=0.5 n=0.5
n=1 n=1
0.5 3.2
n=10 n=10
n=50 n=50
n=1000 2.8 n=1000
0.4
2.4
0.3
2.0
0.2
1.6
0.1 1.2
0.8
0.0
0 10 20 30 40 50 60 100 200 300 400 500 600
(a) Cylindrical shell-clamped (5-60N/m2) (b) Simply supported boundary condition (100-600N/m2)
2.0 n=0
n=0.5
n=1
n=10
1.6
n=50
n=1000
1.2
0.8
0.4
100 200 300 400 500 600

(a) Cylindrical shell-clamped (5-60N/m2)
Figure 4.57 Non-linear load-displacement curves of FGM cylindrical shells for different
magnitude of loading and boundary conditions
248
0.20

0.16
Rx/a=5
0.12
Rx/a=10
Rx/a=20
0.08 Rx/a=50
Rx/a=100
Rx/a=200
0.04
0.00
0 20 40 60 80 100
Volumefraction index (n)
Figure 4.58 Influence of radius on non-linear deflection response of clamped FGM cylindrical
shells for different ranges of volume fraction index (n)
n=0
0.6 n=0.2 n=0
0.6
n=0.5 n=0.2
n=1 n=0.5
0.4 n=1
n=2 0.4
n=1000 n=2
n=1000
0.2
0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
-0.04 0.00 0.04 0.08 0.12 -0.04 0.00 0.04 0.08 0.12
(b) Simply supported boundary condition (h=0.01m, P=1.0x108N/m )

2
(a) Clamped (h=0.01m, P=1.0x106N/m )
2
0.6 0.6
Ceramic
n=0.2
0.4 n=0.5 0.4
n=1
n=2
0.2 Metal 0.2
0.0 0.0
n=0
n=0.2
-0.2 -0.2 n=0.5
n=1
n=2
-0.4 -0.4 n=1000
-0.6 -0.6
-0.01 0.00 0.01 0.02 0.03 0.04 0.001 0.002 0.003 0.004 0.005 0.006 0.007
(c) Clamped boundary condition (h=0.01m, P=1.0x108N/m )(d) Clamped boundary condition (h=0.1m, P=1x10 N/m )
2 8 2
Figure 4.59 Through-the-thickness variation of in-plane axial stress (𝛔xx) of FGM

cylindrical shells for different boundary conditions and thickness values under non-linear
bending
249
4.8.2.3 Geometrically non-linear analysis of FGM spherical shells
n=0
0.48 n=0
n=0.5 3.2
n=0.5
n=1
n=1

n=10 2.8
0.40 n=10
n=50
n=50
n=1000 n=1000
2.4
0.32
2.0
0.24
1.6
0.16
1.2
0.08
0.8
0.00
0 10 20 30 40 50 60 100 200 300 400 500 600
(a) Spherical shell- clamped (5-60N/m2) (a) Spherical shell- simply supported (100-60N/m2)
n=0
n=0.5
n=1
1.6
n=10
n=50
n=1000
1.2
0.8
0.4
100 200 300 400 500 600

(b) Spherical shell- clamped (100-600N/m2)
Figure 4.60 Non-linear load-displacement curves of FGM spherical shells for

different magnitude of loadings and boundary conditions
0.20
0.16
0.12
0.08
Rx/a=5, Rx/a=10
Rx/a=20, Rx/a=50
Rx/a=100, Rx/a=200
0.04
0.00
0 20 40 60 80 100
Figure 4.61 Influence of radius on non-linear deflection response of clamped FGM

spherical shells for different ranges of volume fraction index (n)
250
0.6
n=0 0.6
n=0.2
n=0.5
0.4 n=1 0.4

n=2
n=1000
0.2 0.2
n=0
0.0 0.0 n=0.2
n=0.5
n=1
-0.2 -0.2 n=2
n=1000
-0.4 -0.4
-0.6 -0.6
-0.04 0.00 0.04 0.08 -0.01 0.00 0.01 0.02 0.03 0.04
(a) Simply supported boundary condition
8 2
(b) Clamped boundary condition
(h=0.01m, P=1x10 N/m ) 8 2
(h=0.01m, P=1x10 N/m )
0.6
0.4
0.2
0.0 n=0
n=0.2
n=0.5
-0.2 n=1
n=2
-0.4
n=1000
-0.6
0.004 0.008 0.012 0.016

(a) Clamped boundary condition
(h=0.1m, P=1.0x108N/m )
2
Figure 4.62 Through-the-thickness variation of in-plane normal stresses (𝛔xx) of FGM

spherical shells for different boundary conditions and thickness value under non-linear
bending
GNL response of FGM spherical shells for different boundary conditions and magnitude
of loading is shown in Figure 4.60. The response of load-deflection curve identical to that
of cylindrical shells can be visualized, except the magnitudes of deflections. Again FGM
shells record intermediate deflection between ceramic and metal shells for all the cases
considered.
Similarly, the influence of radius on deflection is similar to the one observed in
Figure 4.61. When in-plane normal stress variation is plotted through-the-thickness, the
top and bottom of the shell experiences tensile and compressive stresses, respectively
(Figure 4.62). Further, it is observed that the thick shells show divergence variation of in-
plane normal stress compared to thin category shells.
251
4.8.2.4 Geometrically non-linear analysis of FGM hyperbolic paraboloid
In Figure 4.63, the non-linear load-displacement curves for hyperbolic parabolid shell are
presented for different volume fraction index values. Various conclusions pertaining to
the volume fraction index and boundary conditions are similar to those discussed under
spherical shells in sub section 4.8.2.3. Similarly, except the magnitude of loading, the
influence of radius on deflection is same as that of spherical and cylindrical shells (Figure
4.64). In Figure 4.65, through-the-thickness variation of axial stress is given for different
thickness values and boundary conditions. Once again, observations concluded from
spherical and cylindrical shell hold good for hyperbolic paraboloid also.
n=0 3.6
n=0
n=0.2 n=0.2
0.30 n=0.5 n=0.5
n=1 n=1
n=2 3.0 n=2
0.24 n=1000 n=1000
2.4
0.18
0.12 1.8
0.06 1.2
0.00
0 10 20 30 40 50 60 100 200 300 400 500 600
(a) Clamped boundary condition (5-60N/m2) (a)Clamped boundary condition (100-600N/m2)

2.0 n=0
n=0.2
n=0.5
n=1
1.6 n=2
n=1000
1.2
0.8
0.4
100 200 300 400 500 600

(a) Clamped boundary condition (100-600N/m2)
Figure 4.63 Non-linear load-displacement curves of FGM hyperbolic paraboloid shells for
different magnitude of loading and boundary conditions
252
0.09
0.08
0.07 h/Rx=-h/Ry=1/300

h/Rx=-h/Ry=1/500
0.06 h/Rx=-h/Ry=1/750
0.05
h/Rx=-h/Ry=1/1000
0.04
0.03
0.02
0.01
0.00
0 20 40 60 80 100

hyperbolic paraboloid shells for different values of volume fraction index (n)
0.6 0.6
0.4 0.4
0.2 0.2
n=0
n=0 n=0.2
0.0 0.0 n=0.5
n=0.2
n=0.5 n=1
-0.2 n=1 n=2
-0.2
n=2 n=1000
n=1000
-0.4 -0.4
-0.6 -0.6
-0.06 -0.04 -0.02 0.00 0.02 0.04 0.06 0.08 0.10 0.12 -0.01 0.00 0.01 0.02 0.03 0.04
(a) Simply supported (b) Clamped boundary condition
(h=0.01m, P=1.0x108N/m )
2
(h=0.01m, P=1.0x108N/m )
2
0.6
0.4
0.2
0.0
n=0
n=0.2
-0.2 n=0.5
n=1
n=2
-0.4 n=1000
-0.6
0.001 0.002 0.003 0.004 0.005 0.006 0.007

(c) Clamped boundary condition
(h=0.1m, P=1.0x108N/m )
2
Figure 4.65 Through-the-thickness variation of in-plane normal stresses (𝛔xx) of FGM

hyperbolic paraboloid shells
253
4.8.2.5 Geometrically non-linear analysis of FGM hypar shells subjected
to mechanical loading
Non-linear load-displacement curves for FGM hypar shell are presented for different
values of volume fraction index values in Figure 4.66. Various conclusions pertaining to
the volume fraction index and boundary conditions are similar to the one discussed under
cylindrical, spherical, and hyperbolic paraboloid shells in the preceeding sub sections.
2.4
0.7 n=0 n=0
n=0.5 n=0.5
n=1
n=1

0.6
n=10 2.0
n=10
n=50 n=50
0.5 n=1000 n=1000
1.6
0.4
0.3 1.2
0.2
0.8
0.1
0.4
0.0
0 10 20 30 40 50 60 100 200 300 400 500 600
(a) Clamped boundary condition (5-60N/m2) (b) Clamped boundary condition (5-60N/m2)
n=0
n=0.5
3.6 n=1
n=10
n=50
3.0 n=1000
2.4
1.8
1.2
100 200 300 400 500 600

(c) Simply supported boundary condition (100-600N/m2)
Figure 4.66 Non-linear load-displacement curves of FGM hypar shells for different
magnitude of loading and different boundary conditions
Similarly, except the magnitude of loading, the influence of radius on deflection is

same as those for other forms of shell geometries considered in this section (Figure 4.67).
In Figure 4.68, through-the-thickness variation of in-plane normal stresses is shown for
254
different thickness and boundary conditions. Once again, observations concluded from
hyperbolic paraboloid shell hold good for hypar also.
0.20

Rxy=0.05, Rxy=0.1,
0.16
Rxy=0.15, Rxy=0.2,
Rxy=3, Rxy=5,
Rxy=10
0.12
0.08
0.04
0.00
0 20 40 60 80 100

hypar shells for different ranges of volume fraction index (n)
n=0
0.6 n=0.2 n=0
n=0.5 0.6
n=0.2
n=1 n=0.5
0.4
n=2 0.4 n=1

n=1000 n=2
0.2
n=1000
0.2
0.0
0.0
-0.2
-0.2
-0.4 -0.4
-0.6 -0.6
-0.006 0.000 0.006 0.012 0.018 -0.004 0.000 0.004 0.008 0.012

(a) Simply supported boundary condition (h=0.01m) (b) Clamped boundary condition (h=0.01m)
0.6
0.4
0.2
0.0
Ceramic
n=0.2
-0.2 n=0.5
n=1
n=2
-0.4 Metal
-0.6
-0.0016 -0.0008 0.0000 0.0008 0.0016 0.0024 0.0032

(c) Clamped boundary condition (h=0.1m)
Figure 4.68 Through-the-thickness variation of in-plane normal stresses (𝛔xx) of FGM hypar
shells
255
4.8.3 Geometrically non-linear analysis of cylindrical, spherical, hypar
and hyperbolic paraboloid FGM sandwich shells subjected to
mechanical loading
Since the present FE shell formulation incorporates all the three radii of curvature in the
stain part, it becomes simple task to perform GNL analysis for shell having different
geometries. To illustrate this statement, different kinds of shell geometries namely,
cylindrical, spherical, hyperbolic paraboloid and hypar shells are considered under non-
linear bending and various observations on corresponding results are discussed under this
section. To exhibit and outline the effects of thickness scheme (i.e., designation of each
layer thickness from bottom of the panel) on non-linear response of FGSS, six types of
thickness schemes were chosen in the study. Thickness schemes 1-8-1, 1-1-1, 2-1-2, 1-2-
1, 2-1-1, 2-2-1 are incorporated and designated as Type 1, Type 2, Type 3, Type 4, Type
5 and Type 6 FGSS panels, respectively, with acronym CSG (cylindrical shell with
graded core), CSH (cylindrical shell with homogeneous core), SPG (spherical shell with
graded core), SPH (spherical shell with homogeneous core), HYPG (hyperbolic
paraboloid with graded core), HYPH (hyperbolic paraboloid with homogeneous core),
HYG (hypar shell with graded core) and HYH (hypar shell with homogeneous core)
associated with different shell forms.
4.8.3.1 Influence of thickness scheme on non-linear bending response of

FGM cylindrical sandwich shells subjected to mechanical loading
Influence of core part in proportion to face sheets in tracing the non-linear path for
cylindrical sandwich shells has been demonstrated in Figure 4.69 for the loading
magnitude within the range 5-60 N/m2. The load was applied in ten incremental steps to
achieve the desired accuracy of non-linear response. Since graded distribution of properties
are high in core compared to face sheets, the curves representing different material profile
parameters (n) are differentiated out from each other in type 1 CSG. With the reason stated
above, Type 1, Type3 3 and Type 2/Type 5 CSG exhibits a close trend among the different
curves representing various values of material gradient parameters. A more common
observation visualized from the non-linear response of different types of FGSS having
cylindrical shell form is that increase in the gradient index results in higher deflection
irrespective of thickness scheme chosen for the present problem.
256
0.5 0.5
Non-dimensional displacement (w/h)

0.4 0.4
n=0 n=0
n=1 n=1
n=5 n=5
0.3 n=10) 0.3 n=10
n=50
0.2 0.2
0.1 0.1
0.0 0.0
0 10 20 30 40 50 60 0 10 20 30 40 50 60

0.6
(a) Type1-CSG 0.5 (b)Type 2-CSG
0.5 n=0
0.4 n=1
n=1.0
n=5
0.4 n=5.0
n=10
n=10
n=50 0.3
0.3
0.2
0.2
0.1 0.1
0.0 0.0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
(c) Type 3-CSG (d) Type 4-CSG
0.4
0.4
n=0
n=0
n=1
n=1
n=5 0.3 n=5
0.3 n=10
n=10
n=50
n=50
0.2
0.2
0.1
0.1
0.0 0.0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
(e) Type 5-CSG (f) Type 6-CSG
Figure 4.69 Non-linear load-deflection behaviour of FGSS (CSG)
257
0.35
0.30 n=0
0.30 n=1

n=5
n=10
Non-dimensional displacment (w/h)
0.25 0.25
n=50
0.20
0.20 n=0
n=1 0.15
n=5
0.15 n=10 0.10
0.05
0.10
0.00
0.05
-0.05
0.00 -0.10
0 10 20 30 40 50 60 10 20 30 40 50 60

(a) Type 1-CSH (b) Type 2-CSH
0.4 0.30
0.25

0.3 n=0 n=0
n=1 0.20 n=1
n=5 n=5
n=10 0.15 n=10
0.2 n=50 n=50
0.10
0.1 0.05
0.00
0.0
-0.05
-0.1 -0.10
0 10 20 30 40 50 60 0 10 20 30 40 50 60
(c) Type 3-CSH (d) Type 4-CSH
0.35 0.30
n=0
n=0
0.30 n=1 0.25

n=1
n=5
0.25 n=5
n=10 0.20
n=10
n=50
0.20 n=50
0.15
0.15
0.10
0.10
0.05
0.05
0.00
0.00
-0.05 -0.05
-0.10 -0.10
10 20 30 40 50 60 0 10 20 30 40 50 60
Non-dimensional displacement (P)
(d) Type 5-CSH
(f) Type 6-CSH
Figure 4.70 Non-linear load-deflection behaviour of FGSS (CSH)
258
2.0
1.6
n=0
Non-dimensiona displacement (w/h)

n=0
1.6
n=0.5 1.4 n=0.5
n=1 n=1
n=10 n=10
n=100 1.2 n==50
1.2 n=100
1.0
0.8 0.8
0.6
0.4
0.4
100 200 300 400 500 600 100 200 300 400 500 600
Non-dimensional displacement (P) Non-dimensional load parameter (P)
2.0
(a) Type 1-CSG 1.8 (b) Type 2- CSG
n=0
1.6 n=0.5

n=1
n=0 n=10
1.6 1.4
n=0.5 n==50
n=1 n=100
n=10 1.2
n==50
1.2 n=100
1.0
0.8
0.8
0.6
0.4
100 200 300 400 500 600 100 200 300 400 500 600

1.6 (c) Type 3-CSG 1.6
(d)Type 4-CSG
n=0
n=0.5 n=0
1.4 1.4
n=1 n=0.5
n=10 n=1
n==50 n=10
1.2 1.2
n=100 n==50
n=100
1.0 1.0
0.8 0.8
0.6 0.6
0.4 0.4
100 200 300 400 500 600 100 200 300 400 500 600
(f) Type 6-CSG
(e) Type 5- CSH
Figure 4.71 Non-linear load-deflection equilibrium path of FGSS (CSG) (CCCC boundary)
259
1.4
1.2 1.2

1.0
0.8 0.8
n=0
n=0.5
n=0 0.6 n=1
n=1 n=10
n=10 0.4 n=50
0.4
n=100 n=100
0.2
0.0 0.0
100 200 300 400 500 600 100 200 300 400 500 600

1.4 (a) Type 1-CSH (b) Type 2-CSH
1.2
1.2
1.0
1.0
n=0 0.8
0.8 n=0.5 n=0
n=1 n=0.5
0.6
0.6
n=10 n=1
n=50 n=10
n=100 0.4
n=50
0.4
n=100
0.2
0.2
0.0
0.0
-0.2
100 200 300 400 500 600 100 200 300 400 500 600

1.4 (c) Type 3-CSH (d) Type 4-CSH
1.6
1.2
1.2
1.0
0.8 n=0
n=0 0.8
n=0.5
n=0.5
n=1
0.6 n=1
n=10
n=10
0.4 n=50
n=50
0.4 n=100
n=100
0.2
0.0
0.0
-0.4
100 200 300 400 500 600 100 200 300 400 500 600
Non-dimensional load paramter (P)
(e) Type 5-CSH
(f) Type 6-CSH
Figure 4.72 Non-linear load-deflection equilibrium path of FGSS (CSH) (CCCC boundary)
Further, a close observation in Figure 4.69 confirms the higher magnitude of displacement
for Type 3 CSG and obviously, the least magnitude for Type 1 CSG. The deflection
response becomes least significant if the value of gradient parameter is above five.
260
On contrary to the FGSS model having graded core, an analogous model with
homogeneous core reflects an interesting non-linear equilibrium path for different
thickness schemes (Figure 4.70). Amplifying the gradient parameter results in decrease of
displacement parameter for various cases and this fact is obvious, since the higher value of
gradient parameter implies the portion of bulky concentration of ceramic segment which is
constituted of higher elastic modulus.
When the gradient parameter assumes the value equals to or above 5, the dissimilar
graphs depicts quite interesting observations for each case. For instance, Type 2 and Type
6 CSG shows almost identical trend compared to other categories of thickness schemes.
Once again, Type 3 CSH gives higher deflection magnitude compared to other cases, by
virtue of lower contribution of ceramic part with respect to higher involvement of ceramic
in face sheet layers for other cases. The effect of magnitude of loading in causing the non-
linear load-displacement response of FGSS with graded and homogenous core is observed
in Figures 4.71 and 4.72. The curves shown for different gradient parameters are highly
non-linear unlike previous cases. However, the conclusions arrived regarding the
maximum and minimum responses for different shell cases are identical to that of previous
cases (Figure 4.69 and 4.70). Under non-linear loading range, different gradient parameters
exhibit dissimilar trend compared to lower magnitudes of loading. For illustration, under
lower magnitude of load, Type 4 CSG shows quite different behavior compared to higher
magnitudes of loadings.
4.8.3.2 Influence of curvature (1/Rx) on non-linear bending of cylindrical

FGSP subjected to mechanical loading
To sketch the influence of curvature on non-linear response of FGSS having graded core,
an illustration has been shown in Figure 4.73 for several values of radii of curvatures
(1/Rx). A loading magnitude of 1x108 N/m2 is applied. The lower value of curvature results
in higher displacement and substantial divergence in response is noticed for higher
curvatures. For instance, Rx/a=100 gives negligible displacement and it may stated that this
statement is true for n value greater than 5. The membrane action is probably the
contributing reason for such type of non-linear responses. Further, the gradient parameter
index greater than 10 will not exhibit any significant response for different thickness
schemes under consideration and the response corresponding to values lesser than 10
shows steep variation of displacement parameter for Type1, Type 2, Type 4 and Type 6
261
CSGs. The curves corresponding to Rx/a=50 and Rx/a=100 resembles close ranges for
non-linear displacement and this is observed irrespective of the type of thickness schemes.
0.16
0.16

0.12
0.12
0.08 Rx/a=5, Rx/a=20,

Rx/a=50, Rx/a=100 0.08
Rx/a=5, Rx/a=20,
Rx/a=50, Rx/a=100
0.04
0.04
0.00 0.00
0 20 40 60 80 100 0 20 40 60 80 100

(a) Type 1-CSG 0.16 (b) Type 2-CSG
0.20
0.14
Rx/a=5, Rx/a=20,
0.16 0.12
Rx/a=50, Rx/a=100
0.10
0.12
0.08 Rx/a=5, Rx/a=20,
Rx/a=50, Rx/a=100
0.08 0.06
0.04
0.04
0.02
0.00
0.00
0 20 40 60 80 100
0 20 40 60 80 100

(c) Type 3-CSG (d) Type 4-CSG
0.16 0.16
Non-diemsnional displacement (w/h)
0.12 0.12
0.08
0.08
Rx/a=5, Rx/a=20,
Rx/a=50, Rx/a=100 Rx/a=5, Rx/a=20,
Rx/a=50, Rx/a=100
0.04
0.04
0.00
0.00
0 20 40 60 80 100 0 20 40 60 80 100

(e) Type 5-CSG (f) Type 6-CSG
Figure 4.73 Influence of curvature on non-linear displacement for CSG (CCCC boundary)
262
4.8.3.3 Variation of in-plane normal stresses (𝛔xx) in cylindrical FGSS
subjected to non-linear bending
n=0
0.6
n=1 0.6 n=0
n=2 n=1

n=5 n=2
0.4 0.4
n=10 n=5
n=10
0.2 0.2
0.0
0.0
-0.2
-0.2
-0.4
-0.4
-0.6
-0.01 0.00 0.01 0.02 0.03 0.04 0.05 -0.02 0.00 0.02 0.04 0.06 0.08
(b) Type 2- CSG
0.6 (a) Type 1- CSG 0.6
n=0
n=0 n=1
0.4
n=1 n=2
0.4
n=2
n=5
n=5 n=10
0.2 n=10 0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
-0.01 0.00 0.01 0.02 0.03 0.04 0.05 0.06 -0.01 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07
(c) Type 3- CSG (d) Type 4- CSG
0.6 0.6
n=0 n=0
n=1 n=1
n=2 0.4 n=2
0.4
n=5 n=5
n=10 n=10
0.2 0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
-0.01 0.00 0.01 0.02 0.03 0.04 0.05 0.06 -0.01 0.00 0.01 0.02 0.03 0.04 0.05 0.06
(e) Type 5- CSG (c) Type 3- CSG
Figure 4.74 Through-the-thickness in-plane stress (𝛔xx) variation of CSG (CCCC boundary)
263
0.4 0.6

0.4
0.2
0.2
0.0 n=0
n=1 0.0
n=2
n=0
n=5
-0.2 -0.2 n=1
n=10
n=2
n=5
-0.4 n=10
-0.4
-0.6
-0.005 0.000 0.005 0.010 0.015 0.020 0.025 -0.02 -0.01 0.00 0.01 0.02 0.03 0.04
(a) Type 1- CSH (b) Type 2- CSH

0.6 0.6
0.4 0.4
0.2 0.2
0.0 n=0 0.0

n=0
n=1 n=1
n=2 n=2
-0.2 -0.2
n=5 n=5
n=10 n=10
-0.4 -0.4
-0.6 -0.6
-0.02 -0.01 0.00 0.01 0.02 0.03 0.04 0.05 -0.02 -0.01 0.00 0.01 0.02 0.03 0.04

(c) Type 3- CSH (d) Type 4- CSH
0.6
0.6
0.4
0.4
0.2
0.2
0.0 n=0 0.0 n=0

n=1 n=1
n=2 n=2
-0.2
n=5 -0.2
n=5
n=10 n=10
-0.4 -0.4
-0.6 -0.6
-0.02 -0.01 0.00 0.01 0.02 0.03 0.04 0.05 -0.01 0.00 0.01 0.02 0.03 0.04
(e) Type 5- CSH (f) Type 6- CSH
Figure 4.75 Through-the-thickness in-plane stress (𝛔xx) variation of CSH (CCCC boundary)
Interesting observations may be visualized when through-the-thickness variation of normal

in-plane stress component was drawn in Figure 4.74 for cylindrical shell under various
thickness schemes. A loading magnitude of 1x108 N/m2 is applied. Except the bottom
regions of panel, most of the shell portion experience tensile stresses for all the cases.
264
Magnitudes of in-plane stresses tend to increase in direct proportion to the values of
volume fraction index. But in Type 2 and Type 4 CSGs, various volume fraction index
values does not exhibit any noticeable change at the top and bottom segments of the panel.
Due to the isotropic property of pure ceramic shell, it depicts through-the-thickness linear
variation of normal stress component. For other values of gradient parameter indexes, a
non-linear response of stress component is observed and this can be more easily visualized
for Type 1, Type 3, Type 5 and Type 6 CSGs.
Among various thickness schemes, Type 2 CSG records maximum

tensile/compressive stresses, while other shell types record almost identical characteristics
for stress component. A quite opposite stress response of in-plane normal stress component
can be visualized when the graded core part is replaced with homogeneous part in
sandwich panels (Figure 4.75). In most of the cases, different curves representing in-plane
normal stress show linear response due to the domination of isotropic property compared
to graded part of panel. For certain values of n (n=5 and 10) values of stresses are not
significant and this is more prominent in Type 4 and Type 6 CSGs. In addition, Type 1
CSG shows a diverging trend compared to other types of CSGs and this observation is
obvious due to the presence of larger proportion of isotropic component (ceramic) in the
panel with respect to face sheet volumes. In other cases, due to the dominance of graded
portion at top and bottom, combination of linear and non-linear responses of in-plane stress
component can be observed through the thickness of the panel.
4.8.3.4 Geometrically non-linear response of FGM spherical sandwich

Non-linear load deflection response for FGM spherical sandwich shells having graded
core part and subjected to mechanical loading is presented in Figure 4.76 for six different
types of sandwich shells. Different thickness schemes are chosen to represent the
influence of core thickness with respect to overall thickness of the plate in predicting the
non-linear response.
Among different types of sandwich shells chosen for the present study, Type 4
CSG exhibit higher deflection. The least deflection is observed for Type 1 CSG. Among
all the types of plates isotropic plates (n=0) shows lower magnitude of deflection. From
265
the non-linear deflection graphs, it can be inferred that the thickness of core and choice of
thickness scheme have significant influence on deflection parameter.
0.40 0.40
0.35 0.35

0.30 n=0 0.30 n=0

n=1 n=1
0.25 n=5 0.25
n=5
n=10 n=10
n=50
0.20 0.20
0.15 0.15
0.10 0.10
0.05 0.05
0.00 0.00
0 10 20 30 40 50 60 0 10 20 30 40 50 60
0.40
(a) Type 1-SPG (b) Type 2-SPG
0.35
0.35 n=0
n=0
n=1
0.30 n=1
0.30 n=5
n=5
n=10 n=10
0.25
0.25 n=50 n=50
0.20
0.20
0.15
0.15
0.10
0.10
0.05 0.05
0.00 0.00
0 10 20 30 40 50 60 0 10 20 30 40 50 60

0.35
(c) Type 3-SPG (d) Type 4-SPG
n=0
0.30
0.30
n=0 n=1
n=1 n=5
0.25 n=5 0.25 n=10
n=10 n=50
n=50
0.20 0.20
0.15 0.15
0.10 0.10
0.05 0.05
0.00 0.00
0 10 20 30 40 50 60 0 10 20 30 40 50 60
(e) Type 5-SPG (f) Type 6-SPG
Figure 4.76 Non-linear load-deflection graph for FGSS (SPG) (CCCC boundary)
266
0.24
0.25

n=0
0.20 n=1
0.20 n=5
n=0
n=10
0.16 n=1
n=50
n=5
n=10 0.15
0.12 n=50
0.10
0.08
0.05
0.04
0.00 0.00
0 10 20 30 40 50 60 0 10 20 30 40 50 60
0.25
(a) Type 1-SPH (b) Type 2-SPH
0.30
n=0 Non-dimensional displacement (w/h)
n=1 0.20 n=0

0.25
n=5 n=1
n=10 n=5
0.20 n=50 n=10
0.15
n=50
0.15
0.10
0.10
0.05
0.05
0.00 0.00
0 10 20 30 40 50 60 0 10 20 30 40 50 60
0.25
(c) Type 3-SPH (d) Type 4-SPH
0.30
0.20
0.25 n=0 n=0
n=1 n=1
n=5 n=5
0.20 0.15
n=10 n=10
n=50 n=50
w/h
w/h
0.15
0.10
0.10
0.05
0.05
0.00
0.00
0 10 20 30 40 50 60
0 10 20 30 40 50 60
P Non-dimensional load parameter (P)
(e) Type 5-SPH (f) Type 6-SPH
Figure 4.77 Non-linear load-deflection graph for FGSS (SPH) (CCCC boundary)
When spherical shells with FGM core are replaced with homogeneous core, a
quite different load-deflection response is observed (Figure 4.77). Linear gradual
response is observed for Type SPH shells, while for other shells linear unsteady response
is observed. If n value beyond 10 is chosen, deflection value is significantly reduced for
267
all types of sandwich shells. This nature of response is more pronounced in case of Type
4 and Type 6 SPH. This tendency represents the higher stiffness of the FGM shell and
thus showing lower magnitude of deflection parameter. Hence, it is vital to suitably
assume the thickness schemes and volume fraction index values in order to arrive at
efficient design for FGM sandwich shells.
When simply supported boundary is chosen to plot the load-equilibrium path of
FGM sandwich shells, a quite divergence response to that of clamped boundary is
visualized (Figure 4.78 and Figure 4.79). As expected, sandwich shells having simply
supported boundary show higher magnitude of deflection compared to clamped shells.
Even though, the pattern of load-deflection curves are identical for both simply supported
and clamped boundaries, Type 1 SPG and SPH do not show gradual deflection response,
which is quite contradictory compared to CSG and CSH shells. Conclusions pertaining to
volume fraction index and loading parameter is same as stated in case of clamped shells. In
Figure 4.80, influence of radius on deflection of spherical shell having graded core is
presented. Excepting for the higher magnitude of loading the load-deflection plots are
identical to that of cylindrical sandwich shells. When n value is above 10, stable response
is observed for all the case. Also, for Rx/a=5, the steep gradient of deflection curve is
observed for all the sandwich shells considered in the present study.
Variation of in-plane normal stresses for spherical sandwich shell having graded
and homogeneous core is shown in Figures 4.81 and 4.82, respectively. In Figure 4.81,
since the core part consists of gradation of material properties divergence response among
the different n curves are observed. But in case of top and bottom layers, the material is
isotropic and hence the linear response can be visualized. Further, it can be seen that,
depending upon the core thickness with respect to total thickness of the shell, non-linear
response of the deflection curve is varied. Further, the top of the shell experience tensile
stresses and the bottom of the shell experiences compressive nature of stresses. Ceramic
shells ensures large magnitude of stresses (either compressive or tensile), while metal
shells record lower magnitude of deflection. Shells with graded properties show
intermediate response between ceramic and metal plates for all the cases. When the core
layer of the spherical shells is replaced with homogenous part, quite different stress
response is observed (Figure 4.82). In all the cases, ceramic shell shows linear variation
and shows tensile and compressive nature of stresses at top and bottom, respectively. Since
the core part is isotropic, linear stress variation is observed.
268
2.8
2.8 n=0
n=0 n=0.5

n=0.5 2.4 n=1
2.4 n=1 n=10
n=10 n=50
n=100 n=100
2.0
2.0
1.6 1.6
1.2 1.2
0.8
0.8
100 200 300 400 500 600 100 200 300 400 500 600
2.8 n=0
n=0
n=0.5
2.4
n=0.5
n=1
n=1
2.4 n=10
n=10
n=50
2.0 n=50
n=100
n=100 2.0
1.6 1.6
1.2
1.2
0.8
0.8
100 200 300 400 500 600 100 200 300 400 500 600
2.4 2.4
n=0
n=0.5
n=1
2.0 n=10
2.0
n=50
n=100
1.6
1.6
n=0
n=0.5
n=1 1.2
1.2
n=10
n=50
n=100 0.8
0.8
100 200 300 400 500 600 100 200 300 400 500 600
Figure 4.78 Non-linear load-deflection graph for FGSS (SPG) (SSSS boundary)
269
2.0 2.0

1.6 1.6
1.2 1.2
n=0
0.8 n=0 0.8 n=0.5
n=1 n=1
n=10 n=10
n=100 0.4
n=50
0.4
n=100
0.0 0.0
100 200 300 400 500 600 100 200 300 400 500 600
2.0
2.0
1.6
1.6
n=0 1.2
1.2 n=0.5 n=0
n=1 n=0.5
n=10 n=1
0.8
0.8 n=50 n=10
n=100 n=50
n=100
0.4 0.4
0.0 0.0
100 200 300 400 500 600 100 200 300 400 500 600
2.0
2.0
1.6
1.6
1.2 n=0 1.2

n=0.5 n=0
n=1 n=0.5
n=10 0.8 n=1
0.8
n=50 n=10
n=100 n=50
n=100
0.4 0.4
0.0 0.0
100 200 300 400 500 600 100 200 300 400 500 600
Figure 4.79 Non-linear load-deflection graph for FGSS (SPH) (SSSS boundary)
The top and bottom skin layers are identified by their graded properties and hence exhibit
different stress response for all types of sandwich shells. Maximum stress values (either
compression or tension) are observed for lower values of volume fraction index (n=1.0).
As the value of volume fraction index increases, the stress values are significantly.
270
0.16
0.14
0.14

0.12
Rx/a=5, Rx/a=20, Rx/a=5, Rx/a=20,
0.12
0.10
Rx/a=50, Rx/a=100 Rx/a=50, Rx/a=100
0.10
0.08
0.08
0.06
0.06
0.04
0.04
0.02 0.02
0.00 0.00
0 20 40 60 80 100 0 20 40 60 80 100
0.18
0.14
0.16
Non-dimensional displacement (w/h) 0.12

0.14
Rx/a=5 Rx/a=5, Rx/a=20,
Rx/a=20
0.12
Rx/a=50
0.10 Rx/a=50, Rx/a=100
Rx/a=100
0.10 0.08
0.08
0.06
0.06
0.04
0.04
0.02
0.02
0.00 0.00
0 20 40 60 80 100 0 20 40 60 80 100
0.14 0.12
0.12
0.10
Rx/a=5 Rx/a=5, Rx/a=20,
Rx/a=20 Rx/a=50, Rx/a=100
0.10
Rx/a=50 0.08
Rx/a=100
0.08
0.06
0.06
0.04
0.04
0.02
0.02
0.00 0.00
0 20 40 60 80 100 0 20 40 60 80 100
(e) Type 5-SPG
(f) Type 6-SPG
Figure 4.80 Influence of curvature on non-linear displacement for SPG sandwich shells
(CCCC boundary)
271
0.6 0.6
0.4

0.4
0.2 0.2
n=0
0.0 0.0 n=0
n=1
n=2 n=1
n=5 n=2
-0.2 -0.2
n=10 n=5
n=1000 n=10
-0.4 n=1000
-0.4
-0.6 -0.6
0.00 0.02 0.04 0.06 0.08 0.00 0.02 0.04 0.06 0.08
(a) Type 1- SPG 0.6 n=0 (b) Type 2-SPG
0.6 n=0
n=1
n=1
n=2
n=2 0.4
0.4 n=5
n=5
n=10
n=10
n=1000
0.2
n=1000 0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
-0.02 0.00 0.02 0.04 0.06 0.08 0.10 0.00 0.02 0.04 0.06 0.08
0.6 n=0 n=0
0.6
n=1 n=1
n=2 n=2
0.4 n=5 0.4 n=5
n=10 n=10
n=1000 n=1000
0.2 0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
0.00 0.02 0.04 0.06 0.08 0.00 0.02 0.04 0.06
Figure 4.81 Through-the-thickness in-plane normal stress (𝛔xx) variation for SPG sandwich
shells (CCCC boundary)
272
n=0
0.6 0.6 n=0
n=1
n=1
n=2
n=2
0.4 n=5 0.4 n=5

n=10
n=10
0.2 0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
-0.02 -0.01 0.00 0.01 0.02 0.03 0.04 -0.02 0.00 0.02 0.04
0.6
n=0
0.6 n=1
n=2
n=5
0.4
0.4 n=10
0.2
0.2
0.0
0.0 n=0
n=1
n=2
-0.2
-0.2 n=5
n=10
-0.4
-0.4
-0.6 -0.6
-0.02 0.00 0.02 0.04 -0.02 0.00 0.02 0.04
0.6 0.6
0.4 0.4
0.2 0.2
0.0 n=0 0.0 n=0

n=1 n=1
n=2 n=2
-0.2 -0.2
n=5 n=5
n=10 n=10
-0.4 -0.4
-0.6
-0.6
-0.02 -0.01 0.00 0.01 0.02 0.03 0.04 0.05
-0.02 0.00 0.02 0.04
Figure 4.82 Through-the-thickness in-plane normal stress (𝛔xx) variation for SPH sandwich
shells (CCCC boundary)
273
4.8.3.5 Geometrically non-linear response of FGM hyperbolic
paraboloid sandwich shells subjected to mechanical loading
Non-linear load-deflection equilibrium path traced for hyperbolic paraboloid sandwich
shells subjected to mechanical loading is presented in Figure 4.83. Loading magnitudes
between 100-600 N/m2 are applied on the top of the shell. Hyperbolic paraboloid shells
exhibit similar characteristics as that of spherical shells, except for the magnitude of
deflection. Increase in the value of volume fraction index, reduces the stiffness of the
shell, and hence increases the deflection. Further, choosing the volume fraction index
value above 50 will not have significant effect in reducing the stiffness of the shell.
Deflection response of the FGM shells lies between ceramic and metal plates for all the
types of sandwiches considered. Since Type 1 HYPG represents the large thickness of
core part compared to top and bottom skin layer i.e., 1-8-1, the gradation curves
representing different material properties are spaced out from each other. On contrary, for
Type 3 HYPG shells, core thickness is 1/5 the total thickness and hence different curves
representing the n values are closely spaced. Unlike spherical shell, these types of shells
have the characteristics of smooth and gradual change of deflection curves in non-linear
range. Among different types of sandwich shells considered in the present study, Type 1
HYPG shells records maximum deflection.
In Figure 4.84, the core part is replaced with homogeneous material and the effect
of load on deflection in non-linear zone is presented for different types of sandwich
schemes. Different n values have significant impact on the deflection response of the
FGM shells. When n value is equal to or above 50, a significant reduction in the
deflection parameter is observed and this observation is more clearly visualized in case of
Type 4 and Type 6 HYPG shells. Ceramic shells gives highest deflection parameter
compared to metal and FGM shells. As the value of n increases from zero, the stiffness of
shell also increases and results in lower deflection values. Variation of in-plane normal
stress is shown in Figures 4.85 and 4.86 for hyperbolic paraboloid sandwich shells by
assuming different n values. For all the cases, the top of the shell experiences a tensile
nature of stresses at the top and compressive nature of stresses at the bottom. Since the
core part is modeled with functional properties, a quite different response of the FGM
shells is observed in the central layer. FGM shells corresponding to higher values of n,
give higher value of compressive and tensile stresses.
274
2.0 n=0
1.8
n=0.5 n=0
1.8 n=1 n=0.5

n=10 1.6 n=1
1.6 n=50 n=10
n=100 n=50
1.4 1.4
n=100
1.2 1.2
1.0
1.0
0.8
0.8
0.6
0.4 0.6
100 200 300 400 500 600 100 200 300 400 500 600
2.0 (b) Type 2-HYPG
(a) Type 1-HYPG n=0
n=0
1.8 n=0.5
n=0.5

1.8
n=1 n=1
n=10 n=10
1.6 1.6
n=50 n=50
n=100 n=100
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0.6
0.6 0.4
100 200 300 400 500 600 100 200 300 400 500 600
1.8
n=0 (c) Type 3-HYPG (d) Type 4-HYPG
1.8 n=0
n=0.5 n=0.5
1.6 n=1
1.6
n=1
n=10 n=10
n=50 n=50
1.4
n=100 1.4 n=100
1.2 1.2
1.0 1.0
0.8
0.8
0.6
0.6
0.4
100 200 300 400 500 600 100 200 300 400 500 600
(e) Type 5-HYPG (f) Type 6-HYPG
Figure 4.83 Non-linear load-deflection equilibrium path for FGSS (HYPG) (CCCC
boundary)
275
1.6
1.4 n=0
n=1 1.4

n=0
n=10
1.2
n=50 n=1
1.2
n=1000 n=10
1.0 n=50
1.0
0.8
0.8
0.6
0.6
0.4 0.4
0.2 0.2
0.0 0.0
100 200 300 400 500 600 100 200 300 400 500 600
(a) Type 1-HYPH (b) Type 2-HYPH
1.6
1.4 n=0
1.4 n=0.5
n=1
1.2
1.2 n=10
n=50
1.0
1.0
n=0
0.8
n=0.5 0.8
n=1
n=10 0.6
0.6
n=50
n=100
0.4 0.4
0.2 0.2
0.0 0.0
100 200 300 400 500 600 100 200 300 400 500 600
(c) Type 3-HYPH (d) Type 4-HYPH

1.6
1.4
1.4
1.2
1.2
1.0 1.0
n=0
0.8 n=0.5 0.8 n=0
n=1 n=0.5
0.6 n=10 0.6 n=1
n=50 n=10
n=50
0.4 0.4
0.2 0.2
0.0 0.0
100 200 300 400 500 600 100 200 300 400 500 600
(e) Type 5-HYPH (f) Type 6-HYPH
Figure 4.84 Non-linear load-deflection equilibrium path for FGSS (HYPH) (CCCC
boundary)
276
0.6 0.6

n=0
0.4
n=0 0.4
n=1
n=1 n=2
0.2
n=2
0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
0.00 0.01 0.02 0.03 0.00 0.01 0.02 0.03 0.04
(a) Type 1-HYPG (b) Type 2-HYPG
0.6 0.6 n=0
n=1

n=2
0.4 0.4
n=0
n=1000
n=1
0.2
n=2 0.2
n=1000
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
0.00 0.01 0.02 0.03 0.04 0.00 0.01 0.02 0.03 0.04
0.6
(c) Type 3-HYPG (d) Type 4-HYPG
n=0 0.6
n=0
n=1 n=1
n=2 n=2
0.4 0.4
n=1000 n=1000
0.2 0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
0.00 0.01 0.02 0.03 0.04 0.00 0.01 0.02 0.03 0.04
(e) Type 5-HYPG (f) Type 6-HYPG
Figure 4.85 Through-the-thickness in-plane stress (𝛔xx) variation for FGSS (HYPG) (CCCC
boundary)
277
n=0
0.6
n=1 n=0
0.6
n=2 n=1
n=2
n=1000
0.4

0.4 n=1000
0.2
0.2
0.0
0.0
-0.2
-0.2
-0.4
-0.4
-0.6
-0.6
-0.10 -0.05 0.00 0.05 0.10 0.15 0.20 0.25
-0.05 0.00 0.05 0.10 0.15 0.20 0.25
In-plane normal stressxx In-plane normal stressxx
(a) Type 1-HYPH (b) Type 2-HYPH
n=0
0.6 0.6
n=1
n=0
n=2
n=1
n=1000
0.4 Non-dimensional depth (z/h) 0.4

n=2
n=1000
0.2 0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
-0.10 -0.05 0.00 0.05 0.10 0.15 0.20 0.25 0.30 -0.05 0.00 0.05 0.10 0.15 0.20 0.25
In-plane normal stressxx In-plane normal stressxx
(c) Type 3-HYPH (d) Type 4-HYPH

n=0
0.6 0.6
n=1
n=2 Cramic
0.4
n=1000 n=1
0.4
n=2
Metal
0.2 0.2
0.0 0.0
-0.2 -0.2
-0.4 -0.4
-0.6 -0.6
-0.05 0.00 0.05 0.10 0.15 0.20 0.25 -0.05 0.00 0.05 0.10 0.15 0.20 0.25
In-plane normal stressxx xx
(e) Type 5-HYPH (f) Type 6-HYPG
Figure 4.86 Through-the-thickness in-plane stress (𝛔xx) variation for FGSS (HYPH) (CCCC
boundary)
278
4.8.3.6 Geometrically non-linear response of hypar sandwich shells
To demonstrate the applicability of the present C0 based formulation to handle the shell of
any kind of geometry; non-linear results are generated for hypar shells with graded and
homogeneous core segments are presented in this sub section. It has been visualized from
Figures 4.86 and 4.87 that, except the higher magnitude of displacement part, hypar shells
depict similar observations as that of cylindrical shells. As observed in previous shell
forms, the role of volume fraction index is to raise the deflection parameter and this
tendency is irrespective of the type of sandwich shell. Further spacing of different
deflection curves representing various n values changes as a function of thickness of core
layer. In all cases, pure isotropic shells (ceramic) records minimum deflection; while shell
representing higher value of n (metal) ensures maximum deflection. FGM shell lies
between ceramic and metal, records intermediate deflection response in all the sandwich
shell schemes. When graded core part is replaced with homogenous core, a quite
interesting and divergence response from that of graded part can be visualized. Except 1-8-
1 type HYG, all other thickness schemes shows identical responses of curves for pure
ceramic (n=0) and with reduced concentration of ceramic part (n=1). This observation
appears to be common for both HYG and HYH panels.
Further, to incorporate and visualize the effect of twist curvature on non-linear

response several values of c/a ratio are considered under different thickness schemes and
the corresponding results are shown in Figure 4.88. The initial steep curve portion
representing increase in deflection parameter is not common for different cases of
sandwich schemes. The slope of such curve varies depending upon the thickness of core
part which is defined by the graded properties. Value of c/a ratio below 3 shows no
remarkable difference in non-linear response of hypar panels having graded core part and
this fact is found to be quite common in all the cases. The observation concerning the
choice of n value for predicting the non-linear response is analogous to that of cylindrical
panels under non-linear bending. Magnitude of tensile stresses in hypar panels are found to
be higher compared to cylindrical shell as observed in Figures 4.89, and different n values
converge at a point at the top segment which is a quite opposite trend visualized at bottom
segment.
279
2.0
n=0 n=0
2.0 n=0.5
1.8

n=0.5
n=1 n=1
n=10 n=10
1.6
n=50 n=50
1.6 n=1000
n=1000
1.4
1.2 1.2
1.0
0.8
0.8
0.4 0.6
100 200 300 400 500 600 100 200 300 400 500 600
2.0 (a) Type 1-HYG (b) Type 2-HYG
n=0 2.0
n=0.5 n=0
n=0.5
1.8 n=1 1.8

n=1
n=10
n=10
n=50
1.6 1.6 n=50
n=1000
n=1000
1.4
1.4
1.2
1.2
1.0
1.0
0.8
0.8
0.6
100 200 300 400 500 600 100 200 300 400 500 600
(c) Type 3-HYG (d) Type 4-HYG
n=0 n=0
1.8 1.8
n=0.5 n=0.5
n=1
n=1
1.6 n=10 1.6 n=10
n=50 n=50
n=1000 n=1000
1.4 1.4
1.2 1.2
1.0 1.0
0.8 0.8
0.6 0.6
100 200 300 400 500 600 100 200 300 400 500 600
(e) Type 5-HYG (f) Type 6-HYG
Figure 4.87 Non-linear load-deflection equilibrium path for FGSS (HYG) (CCCC boundary)
280
1.6 1.6
n=0 n=0
n=1 n=1

n=10
n=10
1.2 n=50 1.2
n=50
n=1000 n=1000
0.8 0.8
0.4 0.4
0.0 0.0
-0.4 -0.4
100 200 300 400 500 600 100 200 300 400 500 600
n=0 (a) Type 1-HYH (b) Type 2-HYH

1.6 n=0
n=1 n=1
n=10 n=10

1.2
n=50 n=50
1.2
n=1000 n=1000
0.8
0.8
0.4
0.4
0.0
0.0
-0.4
100 200 300 400 500 600 100 200 300 400 500 600

(c) Type 3-HYH (d) Type 4-HYH
1.6 1.6
n=0 n=0
n=1 n=1
n=10
n=10
1.2 n=50 1.2 n=50
n=1000 n=1000
0.8
0.8
0.4
0.4
0.0
0.0
-0.4
100 200 300 400 500 600
100 200 300 400 500 600
(e) Type 5-HYH (f) Type 6-HYH
Figure 4.88 Non-linear load-deflection equilibrium path for FGSS (HYH) (CCCC boundary)
281
0.16

0.16
Rxy=0.05, Rxy=0.1,
0.12 Rxy=0.05, Rxy=0.1,
Rxy=0.15, Rxy=0.2,
Rxy=0.15, Rxy=0.2,
Rxy=3, Rxy=5, Rxy=10 0.12 Rxy=3, Rxy=5, Rxy=10
0.08
0.08
0.04
0.04
0.00
0 20 40 60 80 100 0 20 40 60 80 100
Volume fraction index (n) (b)Type 2-HYG
(a)Type 1-HYG
0.16
0.20
Rxy=0.05, Rxy=0.1, Rxy=0.05, Rxy=0.1,

0.16 0.12
Rxy=0.15, Rxy=0.2,
Rxy=0.15, Rxy=0.2,
Rxy=3, Rxy=5, Rxy=10
Rxy=3, Rxy=5, Rxy=10
0.12
0.08
0.08
0.04
0.04
0.00 0.00
0 20 40 60 80 100 0 20 40 60 80 100

(c)Type 3-HYG (d)Type 4-HYG
0.16
0.14
0.12
0.12 Rxy=0.05, Rxy=0.1, Rxy=0.05, Rxy=0.1,
Rxy=0.15, Rxy=0.2, 0.10 Rxy=0.15, Rxy=0.2,
Rxy=3, Rxy=5, Rxy=10 Rxy=3, Rxy=5, Rxy=10
0.08
0.08
0.06
0.04
0.04
0.02
0.00 0.00
0 20 40 60 80 100 0 20 40 60 80 100
(e) Type 5-HYG (f) Type 6-HYG
Figure 4.89 Influence of curvature on non-linear displacement for HYG (CCCC boundary)
282
0.6
n=0
0.5 n=0
n=0.2 0.5
n=0.2
n=0.5
0.4 n=0.5

n=1 0.4 n=1
n=2 n=1000
0.3
0.3 Type 2-HYG
0.2
0.2
0.1
0.0 0.1
-0.1 0.0
-0.2
-0.1
-0.3
-0.2
-0.015 -0.010 -0.005 0.000 0.005 0.010 -0.015 -0.010 -0.005 0.000 0.005 0.010 0.015 0.020
(a) Type 1-HYG (b) Type 2-HYG
0.6
n=0
0.5 n=0.2
n=0.5
0.4 n=1
n=2
0.3
0.2
0.1
0.0
-0.1
-0.2
-0.3
-0.015 -0.010 -0.005 0.000 0.005 0.010 0.015
(c) Type 4-HYG
Figure 4.90 Through-the-thickness in-plane normal stress (𝛔xx) variation for HYG (CCCC
boundary)
4.9 POST BUCKLING ANALYSIS OF FGSP

In this section, to show the application of non-linear FEM approach based on
HSDT developed b for tracing the post buckling equilibrium path of FGSPs, a ceramic-
metal FGP that consists of a combination of Zirconia (ZrO2) and Titanium-Aluminium
oxide (Ti-4V-Al2O3) constituents is considered. A FGSP with geometric properties a=b=1,
and a/h=0.2 is analysed with simply supported boundary condition. Post buckling paths are
traced for FGSPs under different thickness schemes and the results are shown in the form
2
of tables and figures. The non-dimensional parameter N b  N b a is used to evaluate the
Ec h3
buckling loads in primary and post buckled regions.
283
4.9.1 Validation of post buckling analysis of FGSP
To verify the validity of the present FE approach for a single layer FGP, a numerical
problem dealing with post buckling analysis of Zirconia/Aluminium plate is taken from the
literature (Lal et al. 2013) and the comparison of results between the two studies is
depicted in Figure 4.90. At initial value of w/h=0.2, the present results based on HSDT
appears to be closer with the published results and due to the error accumulating at each
iteration step, it shows a large variation when the amplitude ratio attains the maximum
value. Maximum discrepancy between the results occurs for the combination of amplitude
ratio w/h=0.8 and n=0.0 (pure ceramic plate). Even though, both the papers incorporated
the HSDT model with C0 formulation in the frame work of FEM, the constant variation of
transverse displacement was assumed in case of reference paper (Lal et al. 2013) with
seven numbers of nodal unknowns at each node while in the present study variation of
nodal transverse displacement is considered.
Lal et al. (2013)(n=0), Present (n=0),

8 Lal et al. (2013)(n=2), Present (n=2),
Non-dimensional post buckling strength (Nb)
Lal et al. (2013)(n=1000), Present (n=1000)
0.2 0.3 0.4 0.5 0.6 0.7 0.8

Amplitude ratio (w/h)
Figure 4.91 Post buckling response of Al/ZrO2 plate under bi-axial compression (a/h=40,
a/b=1, SSSS boundary)
4.9.2 Non-dimensional critical buckling load of FGSP for different

amplitude ratios
A FGSP with two configurations of core part (FGSP-G and FGSP-H) has been
considered in the subsequent numerical illustrations. Under each category, four kinds of
284
thickness schemes 2-1-2, 1-1-1, 1-2-1 and 1-8-1 are considered and for the sake of
simplicity each scheme is represented as Type A, Type B, Type C and Type D FGSP
(FGSP-G/ FGSP-H), respectively. Different thickness schemes under each category of
FGSP have been selected to reveal the effect of large/small variation of core/face sheet
thickness with respect to overall thickness of the plate. Such a selection will provide a
brief idea regarding the choice of thickness scheme to be chosen with the option of
homogeneous/ graded core to ensure optimum design of FGSPs under in-plane loads.
The load ratio (Nb/Ncr) which is defined as the ratio between primary and post
buckling strength of FGSPs (FGSP-G and FGSP-H) considering various core thickness
values with respect to the total thickness of plate are presented in Tables 4.89 and 4.90 for
different amplitude ratios. In both the tables, different types of FGSPs are arranged in
ascending order from the left in terms of their magnitude of load ratio. A detailed study of
results furnished in Table 4.89 reveals two important observations regarding the post
buckling behavior of FGSPs. First, when the amplitude ratio increases the magnitude of
load ratio increases and this fact is obvious, because at each iteration step the eigen vectors
obtained are scaled up to a desired value and hence get magnified after each iteration
process.
Further, among the different types of FGSPs considered in the example, highest
and lowest value of primary bifurcation strength is observed for Type 4 and Type 1 FGSP-
G, respectively. The reason attributed to the above observation is that when the option of
type 4 FGSP-G is chosen, the core part is almost 8 times more than the thickness of its
counterpart face sheets and hence leads to stiffer plate model.
Table 4.89: Load ratio (Nb/Ncr) of FGSP for different thickness schemes and amplitude ratios
(a/h=100)
Amplitude Nb/Ncr
ratio FGSP-G FGSP-H
(w/h) Type 1 Type 2 Type 3 Type 4 Type 1 Type 2 Type 3 Type 4
2.3999 2.8127 4.1008 5.0940 1.7418 2.4995 3.3224 4.5936
(Ncr) (Ncr) (Ncr) (Ncr) (Ncr) (Ncr) (Ncr) (Ncr)
0.2 2.4178 3.2992 1.9298 2.2404 1.9178 2.2905 2.0389 1.5270
0.4 2.5940 3.3690 2.0002 2.2586 2.0340 2.3974 2.1032 1.5789
0.6 2.8442 3.4524 2.1123 2.3073 2.1771 2.4943 2.1802 1.6422
0.8 3.1303 3.5991 2.2461 2.3424 2.4246 2.5998 2.2737 1.7321
1.0 3.4219 3.7378 2.4167 2.3799 2.6935 2.7179 2.4150 1.8345
1.2 3.8561 3.8839 2.6632 2.4366 2.9904 2.8743 2.6159 1.9401
1.4 5.0883 4.2391 3.1204 2.6546 3.4394 3.1122 2.7755 2.0558
1.6 5.6294 4.4634 3.3723 2.8489 3.7451 3.2194 3.0135 2.1986
1.8 6.1322 4.6617 3.7162 3.0473 4.2675 3.4380 3.2369 2.4190
2.0 2.4178 3.2992 1.9298 2.2404 5.0262 3.6901 3.4412 2.6830
285
Table 4.90: Load ratio (Nb/Ncr) of FGSP for different thickness schemes and amplitude ratios
(a/h=5)
Nb/Ncr
FGSP-G FGSP-H
Amplitude
Type 1 Type 2 Type 3 Type 4 Type 1 Type 2 Type 3 Type 4
ratio
(w/h) 1.7447 2.1460 2.2881 2.2992 2.0206 2.2329 2.2332 2.2554
(Ncr) (Ncr) (Ncr) (Ncr) (Ncr) (Ncr) (Ncr) (Ncr)
0.2 1.6903 1.1627 1.0182 1.0632 1.0987 1.4492 1.1998 1.0597
0.4 1.7004 1.3810 1.0838 1.1456 1.2483 1.4775 1.3197 1.1447
0.6 1.7839 1.0457 1.1629 1.2328 1.3968 1.5745 1.4756 1.3122
0.8 1.9042 1.2175 1.2528 1.3580 1.6313 1.6970 1.7124 1.4429
1.0 2.0608 1.3532 1.3852 1.5072 1.8131 1.8417 1.9220 1.6061
1.2 2.2110 1.4521 1.5845 1.6624 1.9758 1.9509 2.1942 1.7838
1.4 2.2991 1.9679 1.7477 1.8369 2.0907 2.0889 2.3877 2.0179
1.6 2.4831 2.4561 2.2828 2.0683 2.3920 2.2046 2.6767 2.1514
1.8 2.6442 1.3564 2.6217 2.2838 2.6383 2.3537 3.2234 2.3554
2.0 2.8231 4.0961 2.8982 2.5451 3.4259 2.4865 3.8402 2.5961
Similarly, the choice of Type 1 FGSP-G reveals the point that the core part is almost ¼ of
the total thickness i.e., majority of the contribution for post buckling strength is governed
by pure isotropic plates (ceramic and metal). Since the term Ncr (primary buckling
strength) appears in the denominator part of load ratio definition, the values of load ratio
are appears to be higher for Type 1 FGSP-G.
Even though Type-1 FGSP-H is modeled with maximum thickness of skin layers
(face sheets) with respect to total/core thickness, least value of primary bifurcation
strength is observed once again as shown in second part of Table 4.89. Pure ceramic part
plays vital role in this case i.e., it contributes more stiffness to the plate model under
consideration and thus records maximum post buckling strength for plate having large
proportion of core part (Type 4 FGSP-H). Similar observations can be made from Table
4.90 also, except the magnitude of load ratio is lower in this case. As the thickness of the
plate increases the post buckling strength of FGSP-G and FGSP-H are reduces due to the
presence of thickness term in the denominator part of non-dimensional parameter.
286
4.9.3 Influence of volume fraction index (n) on post buckling strengths
of FGSP
Influence of volume fraction parameter on post buckling strength of FGSP having graded
core is shown in Figure 4.91 by considering different values of n. The value of n is chosen
to represent the isotropic (ceramic/metal) and graded plate cases. As expected, high
stiffness property offered by ceramic part ensures greater post buckling strength for
Zirconia plates compared to FGP and Titanium-Aluminium oxide plates. Further, as the
value of n reduces to represent higher concentration of metal component, the post buckling
strength of the plate decreases considerably. Conclusion obtained regarding the superiority
of thickness scheme in terms of post buckling strength is identical to those observed in
previous tables. Since the large volume of FGSP is occupied by graded distribution of
material in Type 4 FGSP-G (1-8-1 thickness scheme), the post buckling curves
representing different graded values (n) are spaced out from each other which is observed
to be an quite opposite trend compared to Type 2 FGSP-G (1-1-1 thickness scheme).
Further an identical trend is observed between Type 2 and Type 3 FGSP-G by recording
intermediate post buckling strength. The effect of replacement of graded core part with the
homogeneous one on plate strength in post buckling region is established in Figure 4.92 for
different types of FGSP-H. Important conclusions pertaining to the effect of volume
fraction parameter and amplitude ratio from Figure 4.91 hold good for Figure 4.92 also,
except the lower values of post buckling parameter are observed in the later case.
4.9.4. Influence of aspect ratio (b/a) on post buckling strengths of FGSP
To show the influence of aspect parameter (b/a) on the post buckling strengths of FGSP-Gs,
Type 2 and Type 4 schemes are picked up with the values of aspect ratio between 1 and 4
(Figure 4.92). The post buckling responses between square plate and rectangular plate with
aspect ratio equal to 1.5 is more pronounced compared to other cases. A close trend is
287
24
15
Non-dimensional post buckling load (Nb)

20 n=0
n=0.5 n=0
n=2 n=0.5
n=10 11 n=2
16
n=1000 n=10
n=1000
12
7
4 3
0
0.0 0.4 0.8 1.2 1.6 2.0 0.0 0.4 0.8 1.2 1.6 2.0
Amplitude ratio (w/h) 8 Amplitude ratio (w/h)
(a) Type 1 FGSP (b) Type 2 FGSP
n=0
14 n=0 n=0.5
n=0.5 n=2
n=2 n=10
n=10 n=1000
10 n=1000 5
2
2
0.0 0.4 0.8 1.2 1.6 2.0 0.0 0.4 0.8 1.2 1.6 2.0
Amplitude ratio (w/h) Amplitude ratio (w/h)
(c) Type 3 FGSP (d) Type 4 FGSP
Figure 4.92 Influence of volume fraction index (n) on post buckling strength of
different FGSP-G for different amplitude ratios (w/h)
5
b/a=1 b/a=1
b/a=1.5
b/a=1.5 4
b/a=2 b/a=2
4 b/a=3 b/a=3
b/a=4 b/a=4
3
3
2
2
1 1
0.0 0.4 0.8 1.2 1.6 2.0 0.0 0.4 0.8 1.2 1.6 2.0
(b)Type 2 FGSP (b)Type 4 FGSP
Figure 4.93 Influence of aspect ratio (b/a) on post buckling strength of different
FGSP-G for different amplitude ratios (w/h)
288
8 b/a=1
b/a=1

b/a=1.5
5 b/a=1.5
7 b/a=2
b/a=2
b/a=3
b/a=3
6 b/a=4
4 b/a=4
5
3
4
3 2
2
1
1
0.0 0.4 0.8 1.2 1.6 2.0 0.0 0.4 0.8 1.2 1.6 2.0
(a)Type 1 FGSP (b)Type 3 FGSP
Figure 4.94 Influence of aspect ratio (b/a) on post buckling strength of different
FGSP-H for different amplitude ratios (w/h)
observed in recording the post buckling strength between the plates with b/a=3.0 and 4.0,
indicating the transformation (more or less) of plate geometry in to a beam. Except the
magnitude of post buckling parameter, the similar observations are made regarding the
effect of aspect ratio on post buckling strength in FGSP-H (Figure 4.94).
4.9.5 Influence of boundary conditions on post buckling strengths of

FGSP
14
12 n=0 (CCSS)
n=0 (CCSS)
n=2 (CCSS)
12 n=2 (CCSS)
n=0 (CSCS)
n=0 (CSCS) 10
n=2 (CSCS)
n=2 (CSCS)
10 n=0 (CSSS)
n=0 (CSSS)
8 n=2 (CSSS)
n=2 (CSSS)
8
6
6
4
4
2 2
0.0 0.4 0.8 1.2 1.6 2.0 0.0 0.4 0.8 1.2 1.6 2.0
Figure 4.95 Influence of boundary conditions on post buckling strength of FGSP-G for
different amplitude ratios (w/h)
289
12
n=0 (CCSS) 10 n=0 (CCSS)
n=2 (CCSS) n=2 (CCSS)

10 n=0 (CSCS) n=0 (CSCS)
n=2 (CSCS) 8 n=2 (CSCS)
8 n=0 (CSSS) n=0 (CSSS)
n=2 (CSSS) 6 n=2 (CSSS)
6
4
4
2
2
0
0.0 0.4 0.8 1.2 1.6 2.0 0.0 0.4 0.8 1.2 1.6 2.0
Figure 4.96 Influence of boundary conditions on post buckling strength of FGSP-H for
different amplitude ratios (w/h)
Three kinds of boundary conditions (CCSS, CSCS, CSSS) are chosen to study their
influences on post buckling strength of FGSP-G and FGSP-H with the combination of
n=0.0 and n=2.0 (Figure 4.95). For all the combinations of boundary conditions and n
values, a ceramic plate ensures higher buckling strength with the choice of CCSS
boundary. As expected, Type 4 FGSP-G has been established to be good choice in post
buckling region. Again, similar observations are made from Figure 4.96 as that of Figure
4.95 regarding the influence of boundary types on post buckling behavior of FGSP-H for
Type 1 and Type 3 thickness schemes.
4.10 SUMMARY
Different numerical examples are solved in this chapter by using the FE formulation
developed in Chapter 3. At first, to ensure the correctness of the formulation, the results
corresponding to isotropic and functionally graded plates/shells are validated with the
results available in the literature. Thereafter, new results are generated for FGM skew
plates/shells by linear/non-linear analysis incorporating various skew angles. Temperature-
dependent material properties are incorporated for the case of thermal analysis. Parametric
studies are performed by considering various material and geometric properties; and the
results are presented in the form of tables and figures. Influences of different material and
geometric properties on static (linear/non-linear), free/forced vibration, and buckling
(linear/non-linear) analyses of FGM plates/shells are discussed briefly in different sections.
290
CHAPTER 5
CLOSURE
5.1 SUMMARY AND CONCLUSIONS
5.1.1 Summary
The various investigations carried out in the present thesis work are summarized as
follows.
 An efficient 2D FE model has been proposed to perform the static, free/forced
vibration, and buckling analyses of FGM plates/shells with/without skew boundary.
The displacement field based on HSDT theory is incorporated in the analysis. The
assumed theory satisfies the shear stress boundary conditions at the top and bottom
of the plate/shell surface and thus eliminates the use of any arbitrary shear
correction factor.
 To address the membrane flexure coupling of FGM structures, an improved
structural kinematics having cubic and constant/quadratic thickness variation in the
in-plane and transverse displacement, respectively is also assumed in the present
research work.
 In the assumed displacement models, the first order derivatives of the transverse
displacement components are substituted by suitable independent field variables to
develop C0 finite element formulation. A nine node isoparametric Lagrangian
element is used in the study which is described by four corner nodes, four mid-side
nodes and one central node. In case of constant transverse displacement variation,
each node is characterized by seven nodal unknowns; while thirteen nodal
unknowns are defined at each node in case of quadratic variation of transverse
displacement.
 In case of analysis of shell structures, the present FE formulation incorporates all
the three radii of curvature (Rx, Ry and Rxy) in the strain part. Hence the formulation
becomes more generalized, where different kinds of shell geometries can be
analyzed.
 Two types of approaches namely, simple Voigt rule of mixture and Mori-Tanaka
schemes are implemented to perform the homogenization. In case of Mori-Tanaka
approach, neighboring interactions among the various parameters are considered.
291
 Non-linear variation of temperature profile is assumed through the thickness and
solution has been obtained by means of solving one-dimensional steady state heat
conduction equation. Also, temperature-dependent material properties of the
constituents are assumed. In all the cases, Young’s modulus and thermal expansion
are considered as temperature and position dependent and thermal conductivity and
density are assumed as independent functions of temperature and dependent only
on position.
 In the present work, a four-parameter dependent power law distribution is assumed
that dictate the various symmetric and asymmetric FGM profiles in the structure.
The gradient parameters exist in the distribution law are suitably assumed to
achieve the different combinations of FGM material profiles.
 In all cases, appropriate values for material gradient parameters are assumed so as
to identify the response of both isotropic and FGM plate/shells under mechanical
and thermal loadings.
 All the research works performed so far for skew FGM plates are based on FSDT
theory. Hence in the present work, to obtain the realistic transverse deformation, a
higher order FGM plate/shell model having skew boundary is considered under
static and dynamic loading conditions. Plate/shell boundary having various skew
angles is analyzed in the present study by means of suitable transformation of
element matrices.
 While analyzing FGM sandwich structures, two types of FGM models are
considered. In the first model, the core portion is modeled by isotropic material and
hence the top and bottom layers are characterized by FGM concept that obeys
simple power law function. In the latter category, the core part is modeled with
FGM material and hence the top and bottom layers are defined by the isotropic
material layers. Various thickness schemes identified by the bottom-core-top layer
thicknesses are incorporated in the numerical examples solved in the study.
Influence of each thickness scheme on various quantities of interest (deflection,
stresses, frequency and buckling loads) is studied in detail using the present FE
formulation.
 Non-linear FE formulation is developed by means of Green-Lagrange strain
equations and strain-displacement relations are arrived by imposing von-Karman
equations. While performing the post buckling analysis of FGM sandwich plates,
292
non-linearity is considered by means of taking the effect of all the non-linear terms
present in the Green-Lagrange strain equations.
5.1.2 Conclusions
Based on the present linear/non-linear FE formulation, many problems are solved related
to the static, vibration, and buckling analyses of FGM plate/shell structures under
mechanical and thermal loading conditions. Various parameters such as aspect ratio,
thickness ratio, radius-side ratio, boundary condition, loading magnitude, volume fraction
index and skew angles are considered and attempt has been made to study the influence of
theses parameters on response of FGM structures.
Based on the various analyses (static, free/forced vibration, buckling, geometric
non-linear and post buckling) performed on FGM plates/shells constituted by single/multi
layer and characterized by single/four-parameter dependent distribution law, the following
imperative conclusions are observed.
 Non-dimensional deflection parameter (w) and axial stresses (σxx) increase as the
volume fraction index (n) rises, and this observation is irrespective of other
geometric parameters (i.e., thickness ratio (a/h) and boundary conditions). But
frequency reduces at higher values of volume fraction index due to the dominance
of stiffness effect of metal constituent.
 As the skew angle of the plate/shell increases, deflection is observed to decrease,
while frequency tends to increase for a given choice of material constituents. When
subjected to thermal load, FGM plate having straight and skew alignment shows
negative value of deflection and records compressive stresses for all the cases of
ceramic and composite plates (Tables 4.7 and 4.8). Under mechanical load, tensile
and compressive stresses are developed at the top and bottom of the plate,
respectively (Table 4.10). Various observations pertaining to influence of
deflection, frequency and stresses for FGM plate geometry, are identical for FGM
shell geometry also, except for the magnitude.
 The critical buckling load parameter of FGM plate increases as the skew angle of
the plate increases and tends to decrease when the constraints are imposed on the
boundaries of the plate. Mori-Tanaka scheme (MTS) predicts higher values of
buckling load in thermal environment (Table 4.33) and Voigt rule of mixture
(VRM) predicts higher values of bucking load under in-plane mechanical loading
(Table 4.35).
293
 An increase in temperature reduces the frequency (Tables 4.22 and 4.23) and
buckling load (Table 4.36) of the FGM plate due to the low value of Young’s
modulus at higher temperatures.
 Hypar skew shells show lower deflection values compared to cylindrical, spherical
and hyperbolic paraboloid shells (Figure 4.1). Under free vibration and dynamic
responses, spherical skew shell establishes better performance compared to
cylindrical skew shell (Figure 4.15) by recording higher frequency values.
 A sharp declining tendency of displacement parameter is identified when the skew
value of the sandwich plate increases (specifically beyond 30) for both Type A
(FGSP with graded core) and Type B (FGSP with homogeneous core) sandwich
plates. i.e., when skew angle is increased from 15 to 30, the displacement
increases about 46% and it is nearly about 75% for 45 sandwich plate.
 Sandwich plates with acute corners (i.e., at skew angle 60 and so) records negative
displacement parameter for both Type A and Type B sandwich plates. This trend is
independent of the magnitude of volume fraction index and various thickness
schemes. Pure ceramic plates show lower value of displacement, while FGM plate
shows higher value of displacement and magnitude of deflection is dependent on
value of volume fraction index.
 Among the various cases of boundary conditions considered (SSSS, CCCC, CFCF
and SCSC), except for the case of CFCF boundary, the influence of volume
fraction index (n) on transverse displacement (w) is insignificant and this effect is
more pronounced in case of Type A sandwich plates than Type B sandwich plates.
 In Type A and Type B sandwich plates, variation of in-plane shear stress (𝜏xy) for
various thickness schemes is significant for skew angle 15 and for further values
of skew angles (30, 45 and 60) no deviations amongst different thickness
schemes are noticed. If in-plane axial stress (xx and yy) variations are plotted for
Type B sandwich plates, pure ceramic plate records minimum stresses (either
compressive or tensile) wherein pure metal plate records maximum stresses (either
compressive or tensile).
 When the influence of curvature-thickness (R/h) ratio is studied on bending
response of functionally graded shells, a sharp rising tendency is observed up to
Rx/a =10.0. An asymptotic response of displacement component is noticed when
R/h assumes the value beyond 10.0.
294
 When the plots of in-plane stress distributions (σxx and σyy) are studied for Type A
FGSSS, it is observed that the magnitude of stresses (both compressive and tensile)
are reduced at higher values of volume fraction indexes. If in-plane and shear stress
variations are studied for Type B FGSSS, interpretations similar to that of Type A
FGSSS are noticed in all cases, except the values of stresses are higher in
magnitude.
 Different FGM models based on four-parameter power law distribution by
incorporating various values of volume fraction indexes show sharp rising trend of
deflection for values, n>5. Further, asymptotic response is ensured when n assumes
the value greater than 50 (Figure 4.20). When aspect ratio is varied from 0.5 to 5, it
is observed that for symmetric profiles, lower values of volume fraction index play
major role in dictating the response of deflection parameter (Figure 4.23). Linear
variation of in-plane normal stress (σxx) is noticed for isotropic and FGM plates and
this trend is not same, when n assumes the value equals 10.0. Marked response was
seen between isotropic and FGM plates in case of variation of in-plane normal
stress (σyy) and this response is more significant at bottom segment (Figure 4.24).
Among different profiles considered in the study, symmetric profiles are better
choice than other profiles by showing lower value of deflection parameter.
 It was noticed that symmetric profiles exhibits maximum frequency value for
different boundary conditions. If one parameter (a1/b1/c1) is varied and two other
parameters are kept constant in four-parameter power law distribution, fall-off
tendency of frequency is observed. Due to the choice of other two parameters in
the power law function, for certain types of modes, FGM plates show frequency
values greater than homogenous ceramic plate.
 Symmetric configuration i.e., FGM2 model shows higher critical buckling load
compared to other FGM (FGM3, FGM4, FGM5 and FGM6) and conventional
models (FGM1). For instance when n=0.2, the critical buckling load difference
between FGM1 and FGM2 model is around 2.57% and it becomes significant as
the n value increases (39.45% at n=10.0) (Figures 4.32 and 4.33).
 If GNL analysis is performed, different sandwich plates shows linear response of
load-deflection curve under lower magnitudes of loading and shows non-linear
load-displacement response for higher loading magnitudes. With the option of
graded core in sandwich plates, Type C plates having 2-1-2 thickness ratio are least
affected by the gradation parameter due to the lower contribution of core part with
295
respect to overall thickness. Non-linear bending response is quite different from
that of FGSP with graded core when the option of graded face sheets has been
chosen. The stiffness effect to minimize the deflection is more pronounced when
the sum of thickness of face sheet layers is reasonably higher compared to
corresponding core portion.
 When post buckling response of FGSP is studied, Type 4 FGSP-G and FGSP-H
show higher post buckling strength among the various thickness schemes
considered. This statement emphasizes the fact that larger volume of graded part
contributes more stiffness to the plate and thus enhancing the post buckling
strength. As far as the post buckling equilibrium paths of FGSP-H are concerned,
the domination of isotropic plates (ceramic) is observed to be significant as
compared to that of graded counterparts. This reflects the stiffness up gradation
offered by pure ceramic plates than that of either graded or pure metal plates. Since
the material gradation in a particular layer is a function of spatial variation of
material properties and in turn the value of parameter ‘n’, additional care has to be
given in selecting this parameter (n).
5.1.3 Major Contributions

The major contributions of the present research works are stated below.
 An efficient and simple 2D FE model incorporating the variation of transverse
displacement component is developed for FGM plate/shell skew structures to
perform static, free vibration and buckling analyses under mechanical and thermal
loading conditions.
 A four-parameter power law distribution is utilized to perform static, free vibration
and buckling of FGM plates based on higher order model. Such a model leads to
various symmetric and asymmetric material profiles in FGM that meet the practical
requirements.
 HSDT model is further extended for the static, free vibration study of sandwich
skew plates/shells for the first time. To perform the task, various thickness schemes
representing either the graded core or homogenous core are studied. Similarly, the
response of isotropic and composites plates are drawn under various values of
volume fraction indexes.
 Geometrically non-linear analysis of functionally graded sandwich plates is
performed for the first time, by incorporating various kinds of sandwich plates
296
under different magnitudes of loading. The various important observations
regarding non-linear bending response sandwich plates may be useful under larger
deformation analysis of FGM sandwich structures.
 Post buckling response of FGM sandwich plates is carried out for the first time by
considering the effect of full non-linearity of Green-Lagrange equations. Such
formulation should accurately capture the non-linear response of FGM plates in
post buckling regions.
5.2 SCOPE OF FUTURE RESEARCH

Following are some of the possible and notable research fields that can be executed as
further extension of the present research work.
 The present non-linear FE formulation can be extended to study of non-linear
free/forced vibration response of FGM skew plates/shells constituted by
single/multilayer concept. Dynamic instability study can be performed for FGM
sandwich plates/shells by considering the skew boundary.
 The present HSDT based finite element formulation can be extended to study the
shape and vibration control of the functionally graded material (FGM) plates with
integrated piezoelectric sensors and actuators. Hence, the influence of the
constituent volume fractions and feedback control gain on the static and dynamic
responses of FGM plates can be examined. Alternatively, a layer of piezoelectric
fiber reinforced composite (PFRC) material can be incorporated in the smart FGM
plate/shell structures. In that case, the effect of variation of piezoelectric fiber angle
in the PFRC layer on its actuating capability of the FG plates/shells can be studied.
 Influence of continuously grading fiber orientation face sheets/core layers on static,
free vibration and buckling of skew sandwich plates/panels can be studied.
 Since most of the studies are based on 1D FGM structures, 2D and 3D FGM
sandwich structures can be analyzed under thermo-mechanical loading conditions.
 FGM plate/shell structures with cutouts of different shapes can be studied using the
present linear/non-linear FE formulation. Thermal post buckling analysis of FGM
skew plates/shells incorporating sandwich part is one of the potential areas of
research in this field.
 FGM structures can be analyzed under moving and impact loadings which are one
of the upcoming research areas.
297
REFERENCES
1. Abdelouahed Tounsi, Mohammed Sid Ahmed Houari, Samir Benyoucef, El

Abbas Adda Bedia. A refined trigonometric shear deformation theory for
thermoelastic bending of functionally graded sandwich plates. Aerospace Science
and Technology, 24(1): 209-220, 2013.
2. Abdullah Sofiyev, Orhan Aksogan, Eckart Schnack, Mehmet Avcar. The
stability of a three-layered composite conical shell containing a FGM layer
subjected to external pressure. Mechanics of Advanced Materials and Structures,
15:461-466, 2008.
3. Aboudi J. Mechanics of composite material: a unified micromechanical approach,
Amsterdam: Elsevier, 1991.
4. Abrate S. Impact on Composite Structures, Cambridge UK: Cambridge University
Press, 1988.
5. Achchhe Lal K.R, Jagtap, Singh B.N. Post buckling response of functionally
graded materials plate subjected to mechanical and thermal loadings with random
material properties. Applied Mathematical Modelling, 37: 2900-2920, 2013.
6. Aggarwal B.D. Bending of parallelogram plates. ASCE Journal of Engineering
Mechanics, 93(4): 9-18, 1967.
7. Ahmed K.N., Mathers M.D. Finite element analysis of anisotropic plates.
International Journal of Numerical Methods in Engineering, 11: 289-307, 1977.
8. Ajay Kumar, Pradeep Bhargava, Anupam Chakrabarti. Vibration of laminated
composite skew hypar shells using higher order theory. Thin-walled structures, 63:
82-90, 2013.
9. Akavci S. Two new hyperbolic shear displacement models for orthotropic laminated
composite plates. Mechanics of Composite Materials, 46(2): 215-226, 2010.
10. Ali J., Bhaskar K., Varadan T. A new theory for accurate thermal/mechanical
flexural analysis of symmetrically laminated plates. Composite Structures, 45: 227-
232, 1999.
11. Aliaga J.W., Reddy J.N. Nonlinear thermoelastic analysis of functionally graded
plates using the third-order shear deformation theory. International Journal of
Computations in Engineering Science, 5:753-779, 2004.
298
12. Alibeigloo A. Thermo elasticity analysis of functionally graded beam with
integrated surface piezoelectric layers. Composite Structures, 92: 1535-1543, 2010.
13. Alijani F. Amabili M., Nonlinear vibrations of thick laminated circular cylindrical
panels. Composite Structures, 96: 643-660, 2013.
14. Alipour M.M., Shariyat M. An elasticity-equilibrium-based zigzag theory for
axisymmetric bending and stress analysis of the functionally graded circular
sandwich plates using a Maclaurin-type series solution. European Journal of
Mechanics A/Solids, 34: 78-101, 2012.
15. Amabili M. A new non-linear higher order shear deformation theory with
thickness variation for large-amplitude vibrations of laminated doubly curved
shells. Journal of Sound and Vibration, 332: 4620-4640, 2013.
16. Amabili M. Nonlinear vibrations of laminated circular cylindrical shells:
comparison of different shell theories. Composite Structures, 94: 207-220, 2011.
17. Amabili M. A non-linear higher-order thickness variation and shear deformation
theory for large-amplitude vibrations of laminated doubly curved shells.
International Journal of Non- Linear Mechanics, 58: 57-75, 2014.
18. Amabili M. Nonlinear vibrations and stability of shells and plates, Cambridge
University Press, New York, 2008.
19. Amabili M., Reddy J.N. A new non-linear higher-order shear deformation theory
for large-amplitude vibrations of laminated doubly curved shells. International
Journal of Non-Linear Mechanics, 45: 409-418, 2010.
20. Anderson T. A. A 3-D elasticity solution for a sandwich composite with a
functionally graded core subjected to transverse loading. In: Proceedings of the
10th US–Japan Conference on Composite Materials, Palo Alto, 82-90, 2002 b.
21. Anderson T.A. Three-dimensional elasticity analysis of sandwich composites with
functionally graded core. In: Proceedings of the 43rd AIAA/ASME/ASCE/AHS/ASC
Structures, Structural Dynamics, and Materials Conference,Denver, 22-25, 2002 a.
22. Ang W.T. A crack in an anisotropic layered material under the action of impact
loading. Journal of Applied Mechanics, 55: 122-125, 1988.
23. Ang W.T., Gurnel A.B. Multiple interacing cracks in an anisotropic multilayered
medium under an antiplane shear stress: a hypersingular integral approach.
Engineering Analysis with Boundary elements, 18: 297-303, 1996.
299
24. Ang W.T., Singh I.V., Tanaka M. An axisymmetric heat conduction model for a
multi-material cylindrical system with application to analysis of carbon nanotube
based composites. International Journal of Engineering Science, 45: 22-33, 2007.
25.Animesh Das, Anjan Dutta, Sudip Talukdar. Efficient dynamic analysis of cable-
stayed bridges under vehicular movement using space and time adaptivity. Finite
Elements in Analysis and Design, 40: 407-424, 2004.
26. Anupam Chakrabarti, Sagar Kr Sengupta, Abdul hamid sheikh. Analysis of
skew composite plates using a new triangular element based on higher order shear
deformation theory. Institution of Engineers India, 85: 77-83, 2004.
27. Arciniega R.A., Reddy J.N. Tensor-based finite element formulation for
geometrically nonlinear analysis of shell structures. Computer Methods in Applied
Mechanics and Engineering, 196: 1048-1073, 2007a.
28. Arciniega R.A., Reddy J.N. Large deformation analysis of functionally graded
shells. International Journal of Solids and Structures, 44: 2036-2052, 2007b.
29. Ashton J.E., Whitney J.M. Theory of laminated plates. In Progress in Material
Science Series. Technomic Publications, Stanford, 4.ASME Applied Mechanics
Reviews, 60:195-216, 2007.
30. Atmane H.A., Tounsi A., Mechab I., Adda Bedia E.A. Free vibration analysis of
functionally graded plates resting on Winkler-Pasternak elastic foundations using a
new shear deformation theory. International Journal Mechanics of Materials and
Design, 6(2): 113-121, 2010.
31. Bahar Uymaz, Metin Aydogdu. Three-dimensional vibration analyses of
functionally graded plates under various boundary conditions. Journal of
Reinforced plastics and composites, 26(18): 1847-1864, 2007.
32. Bahuti A., Eslami M.R., Coupled thermo elasticity of functionally graded
cylindrical shells, Mechanics Research Communications, 34: 1-18, 2007.
33. Bakhti K., Kaci A., Bousahla A.A., Houari M.S.A., Tounsi A., Adda Bedia
E.A. Large deformation analysis for functionally graded carbon nanotubereinforced
composite plates using an efficient and simple refined theory. Steel Composite
Structures, 14(4): 335-347, 2013.
34. Balah M., Al-Ghamedy H.N. Finite element formulation of a third-order
laminated finite rotation shell element. Computers and Structures, 80: 1975-1990,
2002.
300
35. Baskar Y., Ding Y., Schultz R. Refined shear-deformation models for composite
laminates with finite rotations. International Journal of Solids and Structures,
30:2611-2638, 1993.
36. Basset A.B. On the extension and flexural of cylindrical and spherical thin elastic
shells. Philosphical Transactions of the Royal Society of London. Series A,
181:433-480, 1890.
37. Bathe K.J., Ramm E., Wilson E.L. Finite element formulations for large
deformation dynamic analysis. International Journal of Numerical Methods in
Engineering, 9: 353-386, 1975.
38. Belytschko T., Tsay C.S. A stabilization procedure or the quadrilateral plate
element with one point quadrature. International Journal of Numerical Methods in
Engineering, 19: 405-419 1983.
39. Benvensite Y. A new approach to the application of Mori-Tanaka‟s theory in
composite materials. Mechanics of Materials, 6: 147-157, 1987.
40. Benyoucef S., Mechab I., Tounsi A., Fekrar A., Ait Atmane H., Adda Bedia
E.A. Bending of thick functionally graded plates resting on Winkler–Pasternak
elastic foundations. Mechanics of Composite materials, 46(4): 425-434, 2010.
41. Berdichevsky V.L. Nonlinear theory of hard-skin plates and shells. International
Journal of Engineering Science, 48: 357-369, 2010.
42. Bhaskar K., Vardan T.K. Assessment of the self consistent scheme
approximation: conductivity of composites, 68(4): 660-662, 2001.
43. Bich D.H. Nonlinear buckling analysis of FGM shallow spherical shells. Vietnam
Journal of Mechanics, 31: 1-30, 2009.
44. Bich D.H., Dung D.V., Hoa L.K. Nonlinear static and dynamic buckling analysis
of functionally graded shallow spherical shells including temperature effects.
Journal of Composites and Structures, 94 (9): 2952-2960, 2012.
45. Bich D.H., Hoa, L.K. Nonlinear vibration of functionally graded shallow spherical
shell. Vietnam Journal of Mechanics, 32:199-210, 2010.
46. Bich D.H., Tung H.V. Non-linear axisymmetric response of functionally graded
shallow spherical shells under uniform external pressure including temperature
effects. International Journal of Non linear Mechanics, 46: 1195-1204, 2011.
47. Birman V. Stability of functionally graded hybrid composite plates. Composites
Engineering, 5: 913-921, 1995.
301
48. Birman V. Stability of functionally graded shape memory alloy sandwich panels.
Smart Materials and Structures, 6: 278-286, 1997.
49. Birman V., Byrd L.W. Modeling and analysis of functionally graded materials
and structures. Applied Mechanics and Reviews, 60(5), 195-216, 2007.
50. Bodaghi M., Saidi A.R. Levy-type solution for buckling analysis of thick
functionally graded rectangular plates based on the higher-order shear deformation
plate theory. Applied Mathematical Modelling, 34: 3659-3673, 2010.
51. Bodaghi M., Saidi A.R., Stability analysis of functionally graded rectangular
plates under nonlinearly varying in-plane loading resting on elastic foundation.
Archive of Applied Mechanics, 81(6): 765-780, 2011.
52. Bouazza Fahsi, Abdelhakim Kaci, Abdelouahed Tounsi, and El Abbas Adda
Bedia. A four variable refined plate theory for nonlinear cylindrical bending
analysis of functionally graded plates under thermomechanical loadings. Journal of
Mechanical Science and Technology, 26 (12): 4073-4079, 2012.
53. Briassoulis D. Machine locking of degenerated thin shell elements. International
Journal of Numerical Methods in Engineering, 26: 1749-1768, 1988.
54. Brischetto S., Carrera E. Advanced mixed theories for bending analysis of
functionally graded plates. Computers and Structures, 88: 1474-1483, 2010.
55. Brown T. G., Ghali A. Semi-analytical solution of skew box girder bridges.
Proceedings of Institution of Civil Engineers, 59: 487-500, 1975.
56. Brown T. G., Ghali A. Semi-analytical solution of skew plates in bending.
Proceedings of Institution of Civil Engineers, 57: 165-174, 1974.
57. Butalia T.S., Kant T., Dixit V.D. performance of heterosis element for bending of
skew rhombic plates. Computers & Structures, 34(1): 23-49, 1990.
58. Carrera E. C0 Reissner–Mindlin multilayered plate elements including zig–zag
and interlaminar stress continuity. International Journal of Numerical Methods in
Engineering, 39: 1797–820, 1996.
59. Carrera E. Developments, ideas, and evaluations based upon Reissner‟s mixed
variational theorem in the modelling of multilayered plates and shells. Applied
Mechanics and Reviews, 54:301-29, 2001.
60. Carrera E. On the use of murakami‟s zig-zag function in the modeling of layered
plates and shells. Computers and Structures, 82: 541-554, 2004.
302
61. Carrera E. Theories and finite elements for multilayered plates and shells: a unified
compact formulation with numerical assessment and benchmarking. Archives of
Computational Methods in Engineering, 10: 215-296, 2003.
62. Carrera E., Brischetto S., Cinefra M., Soave M. Effects of thickness stretching
in functionally graded plates and shells. Composites: Part B, 42:123-133, 2011a.
63. Carrera E., Brischetto S., Nali P. Plates and shells for smart structures: classical
and advanced theories for modeling and analysis, John Wiley &Sons, UK, 2011b.
64. Caruso J.J., Charnis C.C. Assessment of simplified composite micromechanics
using three dimensional finite element analysis. Journal of Composite Tech Res,
8(3): 77-83, 1986.
65. Castellazzi G., Gentilini C., Krysl P., Elishakoff I. Static analysis of functionally
graded plates using a nodal integrated finite element approach. Composite
Structures, 103: 197-200, 2013.
66. Chamis C.C., Sendeckyj G.P. Critique on theories predicting thermoelastic
properties of fibrous composites, Journal of Composite Materials, 2(3): 332-358,
1968.
67. Chen J.S., Wu C.T., Yoon S., You Y. A stabilized conforming nodal integration
for Galerkin mesh-free methods. International Journal of Numerical Methods in
Engineering, 50: 435-466, 2001.
68. Cheung Y.K., Dashan Z. Large deflection analysis of arbitrary shaped thin plates.
Computers and Structures, 26: 811-814, 1987.
69. Cheung Y.K., Zhou D. Three-dimensional vibration analysis of cantilevered and
completely free isosceles triangular plates. International Journal of Solids and
Structures, 39: 673-687, 2002.
70. Chi S.H, Chung Y.L. Mechanical behavior of functionally graded material plates
under transverse load-Part II: Numerical results. International Journal of Solids
and Structures, 43: 3675-3691, 2006.
71. Cho K.N., Chua L.P., Ghista D.N. Free vibration of laminated rectangular plates
analyzed by higher order individual-layer theory. Journal of sound and Vibration,
45:288-307, 2007.
72. Chung H., Free vibration analysis of circular cylindrical shells. Journal of Sound
and Vibration, 74: 331-350, 1981.
73. Chung Y.L, Chi S.H. The residual stress of functionally graded materials. Journal
of Chinese Institute of Civil Hydraulic Engineering, 13: 1-9, 2001.
303
74. Cinefra M., Carrera E., Brischetto S., Belouettar S. Thermo-Mechanical
Analysis of Functionally graded shells. Journal of Thermal Stresses, 33(10): 942-
963, 2010.
75. Cook R.D, Malkus D.S, Plesha M.E, Concept and Application of Finite Element
Analysis. Third edition, John Wiley and Sons, 1989.
76. Coull A. Stresses in skewed slabs. Concrete, 1: 135-139, 1967.
77. Cowper G. R., Lindburg B. M., Olson M. D. A shallow shell finite element of
triangular shell. International Journal of Solids and Structures, 6: 1133-1156,
1970.
78. Crisfield M.A. Non-linear Finite Element Analysis of Solids and Structures, John
Wiley and Sons, Chichester, England, 2000.
79. Das M., Barut A., Madenci E., Ambur D. R. A triangular plate element for
thermo-elastic analysis of sandwich panels with a functionally graded core.
International Journal of Numerical methods in Engineering, 68: 940-966, 2006.
80. Delale F., Erdogan F. The crack problem for a nonhomogeneous plane. Journal of
Applied Mechanics, 50: 609-614, 1983.
81. Della Croce L., Venini P. Finite elements for functionally graded Reissner-
Mindlin plates. Computer Methods in Applied Mechanics and Engineering, 193(9-
11): 705-725, 2004.
82. Dennis S.T., Palazotto A.N. Large displacement and rotational formulation for
laminated shells including parabolic transverse shear. International Journal of
Non-Linear Mechanics, 25: 67-85, 1990.
83. Desai Y.M., Ramtekkar G.S. and Shah A.H. Dynamic analysis of laminated
composite plates using a layer-wise mixed finite element model. Composite
Structures, 59(2): 237-249, 2003.
84. Donnell L.H. A new theory for the buckling of thin cylinders under axial
compression and bending. Transactions of the ASME, 56: 795-806, 1934.
85. Donnell L.H. Stability of thin-walled tubes under torsion, NACA TR, 479, 1933.
86. Dorman L.H. The thin clamped parallelogram plate under uniform normal
pressure. Department of Supply, Australia, A&L., Rep. SM 214, 1953.
87. Drake J.T, Williamson R.L, Rabin B.H. Finite element analysis of thermal
residual stresses at graded ceramic–metal interfaces Part II: interface optimization
for residual stress reduction. Journal of Applied Physics, 74: 1321-1326, 1993.
304
88. Ebrahimi M.J., Najafizadeh M.M. Free vibration analysis of two-dimensional
functionally graded cylindrical shells. Applied Mathematical Modeling, (38): 308-
324, 2014.
89. Echasz F. L. Structural skew plates, Transactions of ASCE, 111: 101 l-1042, 1946.
90. Eremeyev V., Pietraszkiewicz W. The nonlinear theory of elastic shells with
phase transitions. Journal of Elasticity, 74: 67-86, 2004.
91. Fahsi B., Kaci A., Tounsi A., Adda Bedia E.A. A four variable refined plate
theory for nonlinear cylindrical bending analysis of functionally graded plates
under thermomechanical loadings. Journal of Mechanical Science and Technology,
26(12): 4073-4079, 2012.
92. Farbod Alijani, Marco Amabili. Non-linear vibrations of shells: A literature
review from 2003 to 2013. International Journal of Non-Linear Mechanics, 58:
233-257, 2014.
93. Fares M.E., Elmarghany M.K., Atta D. An efficient and simple refined theory for
bending and vibration of functionally graded plates. Composite Structures, 91(3):
296-305, 2009.
94. Felippa C. A., Bergan P. G. A triangular bending element based on an energy
orthogonal free formulation. Computer Methods in Applied Mechanics and
Engineering, 61: 129-160, 1987.
95.Ferreira A.J.M, Roque C.M.C, Martins P.A.L.S. Analysis of composite plates
using higher-order shear deformation theory and a finite point formulation based on
the multiquadric radial basis function method. Composites: Part B, 34: 627-636,
2003.
96. Ferreira A.J.M. A formulation of the multiquadric radial basis function method for
the analysis of laminated composite plates. Composite structures, 59: 385-392,
2003a.
97. Ferreira A.J.M. Thick composite beam analysis using a global meshless
approximation based on radial basis functions. Mechanics of Advanced Materials
and Structures, 10: 271-284, 2003b.
98. Ferreira A.J.M., Batra R.C., Rouque C.M.C., Qian L.F., Jorge R.M.N. Natural
frequencies of functionally graded plates by a meshless method. Composite
structures, 75: 593-600, 2006a.
305
99. Ferreira A.J.M., Batra R.C., Rouque, C.M.C., Qian L.F., Martins P.A.L.S.
Static analysis of functionally graded plates using third-order shear deformation
theory and a meshless method. Composite Structures, 69: 449-457, 2005a.
100. Ferreira A.J.M., Carrera E., Cinefra M., Roque C.M.C. Analysis of laminated
doubly-curved shells by a layerwise theory and radial basis functions collocation,
accounting for through-the-thickness deformations. Computational Mechanics, 48:
13-25, 2011a.
101. Ferreira A.J.M., Carrera E., Cinefra M., Roque C.M.C., Polit O. Analysis of
laminated shells by a sinusoidal shear deformation theory and radial basis functions
collocation, accounting for through-the-thickness deformations. Composites: Part
B, 42:1276-1284, 2011b.
102. Ferreira A.J.M., Roque C.M.C, Jorge R.M.N. Analysis of composite and
sandwich plate by trigonometric layer-wise deformation theory and radial basis
function. Journal of Sandwich Structures and Materials, 8: 497-515, 2006b.
103. Ferreira A.J.M., Roque C.M.C., Jorge R.M.N. Analysis of composite plates by
trigonometric shear deformation theory and multiquadrics. Computers and
Structures, 83(27), 2225-2237, 2005b.
104. Francesco Tornabene, Erasmo Viola. Static analysis of functionally graded
doubly-curved shells and panels of revolution. Meccanica, 48:901-930, 2013.
105. Francesco Tornabene. Free vibration analysis of functionally graded conical,
cylindrical shell and annular plate structures with a four-parameter power-law
distribution. Computer Methods in Applied mechanics and Engineering,
198:2911-2935, 2009.
106. Fukui Y. Fundamental investigation of functionally gradient material
manufacturing system using centrifugal force. International Journal of Japan
Society of Mechanical Engineers, Series III, 34:144-148, 1991.
107. Ganapathi M. Dynamic stability characteristics of functionally graded materials
shallow spherical shells. Composite Structures, 79: 338-343, 2007.
108. Ganapathi M., Prakash T., Sundararajan N. Influence of functionally graded
material on buckling of skew plates under mechanical loads. Journal of
Engineering Mechanics, 132, 902-905, 2006.
109. Ganapathi M., Makhecha D. Free vibration analysis of multi-layered composite
laminates based on an accurate higher-order theory. Composites Part B:
Engineering, 32: 535-543, 2001.
306
110. Ganapathi M., Prakash T. Thermal buckling of simply supported functionally
graded skew plates. Composite Structures, 74: 247-250, 2006.
111. Getto H., Ishihara S. Development of the fire retardant door with functionally
gradient wood. In: Proceedings of the fourth international symposium on
functionally graded materials, Japan: Tsukuba City; 1996.
112. Ghugal Y. M., Shimpi R.P. A Review of Refined Shear Deformation Theories of
Isotropic and Anisotropic Laminated Plates. Journal of Reinforced Plastics and
Composites, 21(9): 775-813, 2002.
113. Gibson R.F. Principles of composite material mechanics, McGraw-Hill, 1994.
114. Ginsberg J.H. Large-amplitude forced vibrations of simply supported thin
cylindrical shells. ASME Journal of Applied Mechanics, 40: 471-477, 1973.
115. Gossard M.L., Seide P., Roberts W.M. Thermal buckling of plates. Gradient
Materials, 34:173-80, 1993.
116. Grover N., Maiti D.K., Singh B.N. A new inverse hyperbolic shear deformation
theory for static and buckling analysis of laminated composite and sandwich
plates. Composite Structures, 95: 667-675, 2013.
117. Guoyong Jin, Zhu Su, Shuangxia Shi, Tiangui Ye, Siyang Gao. Three-
dimensional exact solution for the free vibration of arbitrarily thick functionally
graded rectangular plates with general boundary conditions. Composite
Structures, 108: 565-577, 2014.
118. Hadid H.A., Ahuja B.M., Ali B.M. Static analysis of clamped skew plates.
Bridge Structural Engineering 9: 39-64, 1979.
119. Hadj Henni ABDELAZIZ, Hassen Ait ATMANE, Ismail MECHAB,
Lakhdar BOUMIA, Abdelouahed TOUNSI, Adda Bedia El ABBAS. Static
analysis of functionally graded sandwich plates using an efficient and simple
refined theory. Chinese Journal of Aeronautics 24, 434-448, 2011.
120. Hadji L., Atmane H.A., Tounsi A., Mechabi I., Addabedia E.A. Free vibration
of functionally graded sandwich plates using four-variable refined plate theory.
Applied Mathematics and Mechanics, 32(7): 925-942, 2011.
121. Haldar S., Sheikh A.H. Bending analysis of composite folded plates by finite
element method. Finite Elements in Analysis and Design, 47: 477-485, 2011.
122. Haldar S., Sheikh A.H. Free vibration analysis of isotropic and composite folded
plates using a shear flexible element. Finite Elements in Analysis and Design, 42:
208-226, 2005.
307
123. Harden C.T., Ruston K.R. The analysis of four span skew bridge using an
electrical analogue computer. Proceedings of the Institution of the Civil
Engineers, 36: 297-324, 1967.
124. Hasin A., Rosen B.W. The elastic moduli of fiber-reinforced materials. ASME
Journal of Applied Mechanics, 4:223-232, 1964.
125. Hasin Z. Analysis of properties of fiber composites with anisotropic constituents.
ASME Journal of Applied Mechanics, 46: 543-550, 1979.
126. Hasin Z. Assessment of the self consistent scheme approximation: conductivity of
composites. Journal of Composite Materials, 4: 284-300, 1968.
127. Hasin Z. The elastic moduli of heterogeneous materials. ASME Journal of
Applied Mechanics, 29: 143-150, 1962.
128. Hasin Z., Shtrikman S. A variational approach to the theory of elastic behavior
of multiphase materials, Journal of Mechanics of Physics and Solids, 11: 127-
140, 1964.
129. Hildebrand F. B., Reissner E., Thomas G. B. Notes on the foundations of the
theory of small displacements of orthotropic shells. NASA TN-1833, Washington,
DC, 1949.
130. Hill R. A self consistent mechanics of composite materials. Journal of Mechanics
of Physics and Solids, 13: 213-222, 1965.
131. Hinton E., Owen D.R.J. Finite Element Software for Plates and Shells. Pineridge
Press, Swansea, U.K., 1984.
132. Hiroyuki Matsunaga. Free vibration and stability of functionally graded plates
according to a 2-D higher-order deformation theory. Composite Structures, 82:
499-512, 2008.
133. Hohe J., Librescu L. A nonlinear theory for doubly curved anisotropic sandwich
shells with transversely compressible core. International Journal of Solids and
Structures, 40: 1059-1088, 2003.
134. Hosseini Kordkheili S. A., Naghdabadi R. Geometrically non-linear
thermoelastic analysis of functionally graded shells using finite element method.
International Journal of Numerical Methods in Engineering, 72:964-986, 2007.
135. Hosseini-Hashemi S., Fadaee M., Atashipour S.R. A new exact analytical
approach for free vibration of Reissner–Mindlin functionally graded rectangular
plates. International Journal of Mechanical Sciences, 53(1): 11-22, 2011.
308
136. Hosseini-Hashemi S., Rokni Damavandi Taher H., Akhavan H., Omidi M.
Free vibration of functionally graded rectangular plates using first-order shear
deformation plate theory. Applied Mathematical Modelling, 34(5): 1276-1291,
2010.
137. Houlston R. Slater J.E. A summary of experimental results on square plate and
stiffened plates subjected to air-blast loading, 57th Shock and Vibration Bulletin,
2:14-20, 1986.
138. Hu H.T., Tzeng W.L. Buckling analysis of skew laminate plates subjected to
uniaxial in-plane loads. Thin-Walled Structures, 38: 53-77, 2000.
139. Hughes T. J. R., Cohen M. The “Heterosis” finite element for plate bending.
140. Hughes T. J. R., Cohen M., Haroun M. Reduced and selective integration
techniques in finite element analysis of plates. Nuclear Engineering and Design,
46:203-222, 1978.
141. Hughes T. J. R., Tezduvar T. E. Finite elements based upon Mindlin plate
theory with particular reference to the four-noded bilinear isoparametric element.
ASME Journal of Applied Mechanics, 48: 587-596, 1981.
142. Huu-Tai Thai, Dong-Ho Choi. An efficient and simple refined theory for
buckling analysis of functionally graded plates. Applied Mathematical Modelling,
36: 1008-1022, 2012.
143. Huu-Tai Thai, Dong-Ho Choi. Efficient higher-order shear deformation
theories for bending and free vibration analyses of functionally graded plates.
Archive of Applied Mechanics, 83: 1755-1771, 2013a.
144. Huu-Tai Thai, Thuc P. Vo. A new sinusoidal shear deformation theory for
bending, buckling and free vibration of functionally graded plates. Applied
Mathematical Modelling, 37: 3269-3281, 2013.
145. Huu-Tai Thai., Dong-Ho Choi. Finite element formulation of various four
unknown shear deformation theories for functionally graded plates. Finite
Elements in Analysis and Design, 75:50-61, 2013b.
146. Hwang C., Pi W. S. Nonlinear acoustic analysis of plates using the finite element
method. AIAA Journal, 10: 276-281, 1972.
147. Igari T., Notomi A., Tsunoda H., Hida K., Kotoh T., Kunishima S. Material
properties of functionally gradient material for fast breeder reactor. In:
309
Proceedings of the first international symposium on functionally gradient
materials, Japan: Sendai, 209–214, 1990.
148. Iqbal Z., Naeem M.N., Sultana N. Vibration characteristics of FGM circular
cylindrical shells using wave propagation approach. Acta Mechanica, 208: 237-
248, 2009.
149. Iyengar K.T.S.R., Srinivasan R.S. Analysis of orthotropic skew-bridge decks.
Bridge Structural Engineering, 1: 23-37, 1971.
150. Jaberzadeh E., Azhari M., Boroomand B. Thermal buckling of functionally
graded skew and trapezoidal plates with different boundary conditions using the
element-free Galerkin method. European Journal of Mechanics A/Solids, 42: 18-
26, 2013.
151. Javaheri R., Eslami M.R., Buckling of functionally graded plates under in-plane
compressive loading. Journal of Applied Mathematics and Mechanics, 82(4):
277-283, 2002.
152. Jenson V. P. Analysis of skew slabs. University of Illinois, Bul, 39, Serial. No.
332, 1941.
153. Jha D.K, Tarun Kant, Singh R.K. A critical review of recent research on
functionally graded plates. Composite Structures, 96: 833-849, 2012a.
154. Jha D.K., Kant T., Singh R.K. Free vibration response of functionally graded
thick plates with shear and normal deformations effects. Composite Structures,
96: 799-823, 2013a.
155. Jha, D.K., Tarun Kant, Singh R.K. Higher order shear and normal deformation
theory for natural frequency of functionally graded rectangular plates. Nuclear
Engineering and Design, 250: 8-13, 2012b.
156. Jha, D.K., Tarun Kant, Singh, R.K., A critical review of recent research on
functionally graded plates. Composite Structures, 96:833-849, 2013b.
157. Jhaa D.K., Tarun Kant, Srinivas K., Singh R.K. An accurate higher order
displacement model with shear and normal deformations effects for functionally
graded plates. Fusion Engineering and Design, 88: 3199- 3204, 2013c.
158. Jie Yang, Sritawat Kitipornchai, Kim Meow Liew. Nonlinear local bending of
FGM sandwich plates. Journal of Mechanics of Materials and Structures, 3(10):
1977-1992, 2008.
159. Jin Z.H., Batra R.C. R curve and strength behavior of functionally graded
material. Material science and Engineering, A242: 70-67, 1998.
310
160. Jin Z.H., Batra R.C. Some basic fracture mechanics concepts in functionally
graded materials. Journal of Mechanics of Physics and Solids, 44: 1221-1235,
1996.
161. Jin Z.H., Batra R.C. Stress intensity relaxation at the tip of an edge crack in a
functionally graded material subjected to a thermal shock. Journal of thermal
stresses, 19: 317-339, 1996.
162. Kaci A., Bakhti K., Hebali H., Tounsi A. Mathematical solution for nonlinear
cylindrical bending of sigmoid functionally graded plates, Journal of Applied
Mechanics. Technology and Physics, 54 (1): 124-131, 2013.
163. Kant T. Numerical analysis of thick plates. Computer Methods in Applied
Mechanics and Engineering, 31(1):1-18, 1982.
164. Kant T. Pandya B.N. A simple finite element formulation of a higher-order
theory for unsymmetrically laminated composite mates. Composite Structures 9:
215-246, 1988.
165. Kant T., Khare R.K. A higher-order facet quadrilateral composite shell element.
International Journal of Numerical Methods in Engineering, 40: 4477-4499,
1997.
166. Kant T., Kommineni J.R. C0 finite element geometrically non linear analysis of
fibre reinforced composite and sandwich laminates based on a higher order
theory. Computers and Structures, 45(3): 511-520, 1992.
167. Kant T., Manjunatha B.S. On Accurate Estimation of Transverse Stresses in
Multilayer Laminates. Computers and Structures, 50:351-365, 1994.
168. Kant T., Owen D.R.J., Zeinkwicz O.C. A refined higher order C0 plate bending
element. Journal of Computers and Structures, 15: 177-183, 1982.
169. Kant T., Pandya B.N. A simple finite element formulation of a higher-order
theory for unsymmetrically laminated composite plates. Composite Structures,
9(3): 215-246, 1988.
170. Kant T., Pendhari S.S. and Desai Y.M. An efficient semi-analytical model for
composite and sandwich plates subjected to thermal load. Journal of thermal
Stresses, 31: 77-103, 2008.
171. Kant T., Swaminathan K. Analytical solutions for free vibration of laminated
composite and sandwich plates based on a higher order refined theory. Composite
Structures, 53: 73-85, 2001.
311
172. Kant T., Swaminathan K. Estimation of transverse/interlaminar stresses in
laminated composites- A Selective Review and Survey of Current Developments.
Composite Structures, 49:65-75, 2000.
173. Karama M., Afaq K.S., Mistou S. Mechanical behaviour of laminated
composite beam by the new multi-layered laminated composite structures model
with transverse shear stress continuity. International Journal of Solids and
Structures, 40(6): 1525-1546, 2003.
174. Karami Khorramabadi, Najafizadeh M.M., Alibabaei Shahraki J.,
Khazaeinejad P. Effect of shear deformation theories on free vibration of
functionally graded plates. World Academy of Science Engineering and
Technology, 48: 216-221, 2008.
175. Kashtalyan M., Menshykova M. Three-dimensional elasticity solution for
sandwich panels with a functionally graded core. Composite Structures, 87: 36–
43, 2009.
176. Kennedy J. B. Clamped skew plate under uniform normal loading. Aeronautical
Journal of the Royal Aeronautical Society, 72: 338-340, 1968.
177. Kennedy J. B. On bending of clamped skewed plates under uniform pressure.
Aeronautical Journal of the Royal Aeronautical Society, 69: 352-355, 1965.
178. Kennedy J. B., Huggins M. W. Series solution of skewed stiffened plates.
ASCE Journal of Engineering Mechanics, 90: l-22, 1964.
179. Khalili S.M.R., Mohammadi Y. Free vibration analysis of sandwich plates with
functionally graded face sheets and temperature-dependent material properties: A
new approach. European Journal of Mechanics A/Solids, 35: 61-74, 2012.
180. Kim J.I., Kim W.J., Choi D.J., Park J.Y., Ryu W.S. Design of a C/SiC
functionally graded coating for the oxidation protection of C/C composites.
Carbon, 43: 1749-1757, 2005.
181. Kirchhoff, G. Uber das gleichgewicht und die bewegung einer elastischen
scheibe. Journal fur die reine und Angewandte Mathematik, 40(1): 51-88, 1850.
182. Koiter W.T. On the nonlinear theory of thin elastic shells, Proceedings
Koniklijke Nederlands Akademie van Wetenschappen B69, 1-54, 1966.
183. Koizumi M. The concept of FGM. Ceramic transactions of functionally graded
materials, 34: 3-10, 1993.
312
184. Kolar V., Nemec I. The efficient finite element analysis of rectangular and skew
laminated plates. International Journal of Numerical Methods in Engineering, 7:
309-323, 1973.
185. Krishna Reddy A.R., Palaninathan R. Free vibration of skew laminates.
186. Lee Y.D, Erdogan F. Residual/thermal stress in FGM and laminated thermal
barrier coatings. International Journal of Fracture, 69: 145-165, 1995.
187. Lee Y.Y., Zhao X., Liew K.M. Thermoelastic analysis of functionally graded
plates using the element-free kp-Ritz method. Journal of Smart materials and
Structures, 18: 035007-035015, 2009.
188. Lee Y.Y., Zhao X., Reddy J.N. Postbuckling analysis of functionally graded
plates subject to compressive and thermal loads. Computer Methods in Applied
Mechanics and Engineering, 199: 1645-1653, 2010.
189. Li Q., Iu V.P., Kou K.P. Three-dimensional vibration analysis of functionally
graded material sandwich plates. Journal of Sound and Vibration, 311: 498-515,
2008.
190. Liew K.M., Zhao X., Lee Y.Y. Post buckling responses of functionally graded
cylindrical shells under axial compression and thermal loads. Composites: Part B,
43: 1621-1630, 2012.
191. Liu L., Chua L.P., Ghista D.N. Mesh-free radial basis function method for
static, free vibration and buckling analysis of shear deformable composite
laminates. Composite Structures, 78:58-69, 2007.
192. Liu, Li. An overall view of laminate theories based on displacement hypothesis.
Journal of Composite Materials, 30: 1539-1561, 1996.
193. Lo K. H., Christensen R.M. and Wu E.M. A higher order theory of plate
deformation, part 2: Laminated plates. Transactions of ASME, Journal of Applied
Mechanics; 44(4): 669-676, 1977b.
194. Lo K.H., Christensen R.M., Wu E.M. A high order theory of plate
deformation, part-1: Homogeneous plates. Transactions of ASME, Journal of
Applied Mechanics; 44(4): 663-668, 1977a.
195. Loja M.A.R., Mota Soares C.M., Barbosa J.I. Analysis of functionally graded
sandwich plate structures with piezoelectric skins using B-spline finite strip
method. Composite Structures, 96: 606-615, 2013.
313
196. Lorenzo Dozio. Natural frequencies of sandwich plates with FGM core via
variable-kinematic 2-D Ritz models. Composite Structures, 96: 561-568, 2013.
197. Love A.E.H. A treatise on the mathematical theory of elasticity, 4th edition,
Cambridge, Cambridge University Press, 1952.
198. Loy C.T., Lam K.Y., Reddy J.N. Vibration of functionally graded cylindrical
shells. International Journal of Mechanical Sciences, 41: 309-324, 1999.
199. Madhukar S., Singha M.K. Geometrically non linear finite element analysis of
sandwich plates using normal deformation theory. Composite Structures, 97, 84-
90, 2013.
200. Makhecha D., Ganapathi M., Patel B. Dynamic analysis of laminated composite
plates subjected to thermal/mechanical loads using an accurate theory, Composite
Structures, 51: 221-236, 2001.
201. Malekzadeh P., Heydarpour Y. Response of functionally graded cylindrical
shells under moving thermo-mechanical loads. Thin-Walled Structures, 58: 51-
66, 2012.
202. Maloy K. Singha., Ramachandra L.S., Bandyopadhyay J.N. Vibration
behavior of thermally stressed composite skew plate. Journal of Sound and
Vibration, 2296: 1093-1102, 2006.
203. Maloy K.Singha, Ramachandra L.S., Bandyopadhyay J.N. Thermal
postbuckling analysis of laminated composite plates. Composite Structures, 54,
453-458, 2001.
204. Manjunatha B.S., Kant T. A comparison of 9 and 16 node quadrilateral
elements based on higher-order laminate theories for estimation of transverse
stresses. Journal of Reinforced Plastics and Composites, 11(9): 968-1002, 1992.
205. Mantari J.L, Guedes Soares C. Generalized hybrid quasi-3D shear deformation
theory for the static analysis of advanced composite plates. Composite Structures,
94: 2561-2575, 2012 b.
206. Mantari J.L., Guedes Soares C. A novel higher-order shear deformation theory
with stretching effect for functionally graded plates. Composites: Part B, 45: 268-
281, 2013a.
207. Mantari J.L., Guedes Soares C. Bending analysis of thick exponentially graded
plates using a new trigonometric higher order shear deformation theory.
Composite Structures 94: 1991-2000, 2012a.
314
208. Mantari J.L., Guedes Soares C. Finite element formulation of a generalized
higher order shear deformation theory for advanced composite plates. Composite
Structures, 96: 545-553, 2013b.
209. Mantari J.L., Guedes Soares C. Optimized sinusoidal higher order shear
deformation theory for the analysis of functionally graded plates and shells.
Composites: Part B, 56:126-136, 2014.
210. Mantari J.L., Oktem A.S., Guedes Soares C. A new trigonometric shear
deformation theory for isotropic, laminated composite and sandwich plates.
International Journal of Solids and Structures, 49(1):43-53, 2012a.
211.Mantari J.L., Oktem A.S., Guedes Soares C. Bending response of functionally
graded plates by using a new higher order shear deformation theory, Composites
Structures, 94(2): 714-723, 2012b.
212. Matin Latifi, Fatemeh Farhatnia, Mahmoud Kadkhodaei. Buckling analysis
of rectangular functionally graded plates under various edge conditions using
Fourier series expansion. European Journal of Mechanics A/Solids, 41:16-27,
2013.
213. Matsunaga H. Free vibration and stability of functionally graded plates according
to a 2-D higher-order deformation theory. Composite Structures, 82(4): 499-512,
2008.
214. Maziar Janghorban, Amin Zare. Thermal effect on free vibration analysis of
functionally graded arbitrary straight -sided plates with different cutouts, Latin
American Journal of Solids and Structures, 8: 245-257, 2011.
215. Meiche N.E., Tounsi A. Ziane N., Mechab I., Adda Bedia E.A. A new
hyperbolic shear deformation theory for buckling and vibration of functionally
graded sandwich plate. International Journal of Mechanical Sciences, 53(4):
237–247, 2011.
216. Mindlin, R.D., Influence of rotary inertia and shear on flexural motions of
isotropic, elastic plates. Journal of Applied Mechanics, 18(1): 31-38, 1951.
217. Ming P.G., Song L.F. A new element used in the non-orthogonal boundary plate
bending theory-an arbitrary quadrilateral element. International Journal of
Numerical Methods in Engineering, 24: 1031-1042, 1987.
218. Mirsky I. The deflection of thin flat clamped parallelogram plate subjected to
uniform normal loading. Department of Supply, Australia, A. R. L., Rep. SM 175,
1951.
315
219. Miyamoto Y. Functionally graded materials: design, processing and applications,
Dordrecht, Netherlands: Kluwer Academic Publications, 1999.
220. Mohammad Talha, Singh B.N. Static response and free vibration analysis of
FGM plates using higher order shear deformation theory. Applied Mathematical
Modelling, 34: 3991-4011, 2010.
221. Mohammadi M., Saidi A.R., Jomehzadeh E. Levy solution for buckling
analysis of functionally graded rectangular plates. Applications of Composite
Materials, 17(2):81-93, 2010.
222. Mohammed Sid Ahmed Houari, Abdelouahed Tounsi, Anwar Bég O. Thermo
elastic bending analysis of functionally graded sandwich plates using a new
higher order shear and normal deformation theory. International Journal of
Mechanical sciences, 76: 102-111, 2013.
223. Monforton G. R., Michail M. G. Finite element analysis of skew sandwich
plates. ASCE Journal of Engineering Mechanics, 98: 763-769, 1972.
224. Mori T., Tanaka T. Average stress in matrix and average elastic energy of
materials with misfitting inclusion. Acta Metall, 21: 571-574, 1973.
225. Morley L. S. D. Bending of clamped rectilinear plates. Journal of Mechanics and
Applied Mathematics, 27: 293-317, 1964.
226. Morley L. S. D. Bending of clamped rectilinear plates. R.A.E. report on
Structures, 289, 1963.
227. Morley L. S. D. Bending of simply supported rhombic plate under uniform
normal loading. Ministry of Aviation, London, A.R.C., 23, 1961.
228. Morley L. S. D. Bending of simply supported rhombic plate under uniform
normal loading. Journal of Mechanics and Applied Mathematics, XV: 413-426,
1962.
229. Morley L. S. D. Skew Plates and Structures. Institute of Series of Monographs on
Aeronaurics and Astronautics. Pergamon Press, New York, 1963.
230. Moya J.S. Layered ceramics. Advanced Materials, 7:185-189, 1995.
231. Mukhopadhyay M. Finite strip method of analysis of clamped skew plates in
bending. Proceedings of Institution of Civil Engineers, 61: 189-195, 1976.
232. Murakami H. Laminated composite plate theory with improved in-plane
responses. Journal of Applied Mechanics, 53: 661-666, 1986.
316
233. Najafizadeh M. M., Isvandzibaei M. R. Vibration of functionally graded
cylindrical shells based on higher order shear deformation plate theory with ring
support. Acta Mechanica, 191: 75-91, 2007.
234. Najafizadeh M. M., Isvandzibaei M. R.Vibration of functionally graded
cylindrical shells based on different shear deformation shell theories with ring
support under various boundary conditions. Journal of Mechanical Science and
Technology, 23: 2072-2084, 2009.
235. Najafizadeh M.M., Eslami M.R. Buckling analysis of circular plates of
functionally graded materials under uniform radial compression. International
Journal of Mechanical Sciences, 44 2479-2493, 2002.
236. Najafizadeh M.M., Heydari H.R. An exact solution for buckling of functionally
graded circular plates based on higher order shear deformation plate theory under
uniform radial compression. International Journal of Mechanical Sciences,
50(3): 603-612, 2008.
237. Najafizadeh M.M., Heydari H.R. Thermal buckling of functionally graded
circular plates based on higher order shear deformation plate theory. European
Journal of Mechanics A/Solids, 23 1085-1100, 2004.
238. Najafizadeh M.M., Isfandzibaei M.R. Vibration of functionally graded
cylindrical shells based on higher order shear deformation plate theory with ring
support. Acta Mechanica, 191:75-91, 2007.
239. Najafizadeh M.M., Isfandzibaei M.R. Vibration of functionally graded
cylindrical shells based on different shear deformation shell theories with ring
support under various boundary conditions. Journal of Mechanical Science and
Technology, 23: 2072-2084, 2009.
240. Naruoka M., Ohmura H. On the analysis of skew girder bridge by theory of
orthotropic parallelogram plates. Proceedings of IA&SE 19: 231-256, 1959.
241. Natarajan S., Ganapathi Manickam. Bending and vibration of functionally
graded material sandwich plates using an accurate theory. Finite elements in
analysis and design, 57: 32-42, 2012.
242. Nelson R.B. Larch D.R. A refined theory of laminated orthotropic plates.
Journal of Applied Mechanics, 41:177-183, 1974.
243. Nemat-Alla M. Reduction of thermal stresses by developing two dimensional
functionally graded materials. International Journal Solids and Structures, 40:
7339-7356, 2003.
317
244. Neves A.M.A., Ferreira A.J.M., Carrera E., Cinefra M., Roque C.M.C., Jorge
R.M.N. Soares C.M.M. Free vibration analysis of functionally graded shells by a
higher-order shear deformation theory and radial basis functions collocation,
accounting for through-the-thickness deformations. European Journal of
Mechanics A/Solids, 37: 24-34, 2013a.
245. Neves A.M.A., Ferreira A.J.M., Carrera E., Cinefra M., Jorge R.M.N.,
Soares C.M.M. Static analysis of functionally graded sandwich plates according
to a hyperbolic theory considering Zig-Zag and warping effects. Advances in
Engineering Software, 52: 30-43, 2012a.
246. Neves A.M.A., Ferreira A.J.M., Carrera E., Cinefra M., Roque C.M.C.,
Jorge R.M.N., Soares C.M.M. A quasi-3D hyperbolic shear deformation theory
for the static and free vibration analysis of functionally graded plates. Composite
Structures, 94: 1814-1825, 2012b.
247. Neves A.M.A., Ferreira A.J.M., Carrera E., Cinefra M., Roque C.M.C.,
Jorge R.M.N., Soares C.M.M. Static, free vibration and buckling analysis of
isotropic and sandwich functionally graded plates using a quasi-3D higher-order
shear deformation theory and a meshless technique. Composites: Part B, 44: 657-
674, 2013b.
248. Neves A.M.A., Ferreira A.J.M., Carrera E., Roque C.M.C., Cinefra M.,
Jorge R.M.N. Bending of FGM plates by a sinusoidal plate formulation and
collocation with radial basis functions. Mechanics Research and
Communications, 38:368-371, 2011.
249. Neves A.M.A., Ferreira A.J.M., Carrera E., Roque C.M.C., Cinefra M.,
Jorge R.M.N. A quasi-3D sinusoidal shear deformation theory for the static and
free vibration analysis of functionally graded plates. Composites: Part B, 43:711-
25, 2012c.
250. Nguyen Dinh Duc, Vu Thi Thuy Anh, Pham Hong Cong. Nonlinear
axisymmetric response of FGM shallow spherical shells on elastic foundations
under uniform external pressure and temperature. European Journal of
Mechanics A/Solids, 45: 80-89, 2014.
251. Noureddine El Meiche, Abdelouahed Tounsi, Noureddine Ziane, Ismail
Mechab, El Abbes Adda. Bedia. A new hyperbolic shear deformation theory for
buckling and vibration of functionally graded sandwich plate. International
Journal of Mechanical Sciences, 53: 237-247, 2011.
318
252. Novozhilov V.V. Foundations of the Nonlinear Theory of Elasticity, Graylock
press, New York, 1953.
253. Oktem A.S., Mantari J.L., Guedes Soares C. Static response of functionally
graded plates and doubly-curved shells based on a higher order shear deformation
theory. European Journal of Mechanics A/Solids, 36: 163-172., 2012.
254. Ömer Civalek. Free vibration and buckling analyses of composite plates with
straight-sided quadrilateral domain based on DSC approach. Finite Elements in
Analysis and Design 43: 1013 -1022, 2017.
255. Oonishi H., Noda T., Ito S., Kohda A., Yamamoto H., Tsuji E. Effect of
hydroxyapatite coating on bone growth into porous titanium alloy implants under
loaded conditions. J Appl Biomater, 5(1):23-27, 1994.
256. Opoka S. Pietraszkiewicz W. On modified displacement version of the nonlinear
theory of thin shells. International Journal of Solids and Structures, 46: 3103-
3110, 2009.
257. Opoka S., Pietraszkiewicz W. Intrinsic equations for non-linear deformation and
stability of thin elastic shells. International Journal of Solids and Structures, 41:
3275-3292, 2004.
258. Osaka T., Matsubara H., Homma T., Mitamura S., Noda K. Microstructural
study of electroless-plated CoNiReP/NiMoP double layered media for
perpendicular magnetic recording. Journal of Applied Physics, 29(10):1939-1943,
1990.
259. Owen D. R. J., Figueiras J. A. Anisotropic elastoplastic finite element analysis
of thick and thin plates and shells. International Journal of Numerical Methods in
Engineering, 19: 541-566 1983.
260. Padmanav Dash, Singh B.N. Buckling and post-buckling of laminated composite
plates. Mechanics Research Communications, 46: 1-7, 2012.
261. Pagano N.J. Exact solutions for composite laminates in cylindrical bending.
Journal of Composite Materials, 3(3):398-411, 1969.
262. Pagano N.J. Exact solutions for rectangular bidirectional composites and
sandwich plates. Journal of Composite Materials, 4(1):20-34, 1970.
263. Palazotto A.N., Dennis S.T. Nonlinear Analysis of Shell Structures. AIAA
Educational Series, Washington DC, 1992.
319
264. Pan E., Han F. Exact solution for functionally graded and layered magneto-
electro-elastic plates. International Journal of Engineering Science, 43: 321-339,
2005.
265. Panda S.K., Singh B.N. Thermal post-buckling behavior of laminated composite
cylindrical/hyperboloid shallow shell panel using non linear finite element
method. Composite Structures, 91: 366-374, 2009.
266. Pandit M.K. An Improved Plate Model for Static, Vibration and Buckling
Response of Sandwich Laminates Having Random Material Properties. Ph.D.
Thesis, Department of Ocean Engineering and naval Architecture, Indian
Institute of Technology, Kharagpur, 2009.
267. Pandya B.N., Kant T. Finite element stress analysis of laminated composite
plates using higher order displacement model. Composites Science and
Technology, 32:137-155, 1988c.
268. Pandya B.N., Kant T. Flexure analysis of laminated composites using refined
higher-order Co plate bending elements. Computer Methods in Applied
Mechanics and Engineering, 66:173-198, 1988 b.
269. Pandya B.N., Kant T. Higher-order shear deformable theories for flexure of
sandwich plates-finite element evaluations, International Journal of Solids and
Structures, 24(12):1267-1286, 1988 a.
270. Parisch H. A continuum-based shell theory for non-linear applications.
International Journal for Numerical Methods in Engineering, 38: 1855-1883,
1995.
271. Park T., Lee S.Y., Seo J.W., Voyiadjis G.Z. Structural dynamic behavior of
skew sandwich plates with laminated composite faces. Composites: Part B, 39:
316-326, 2008.
272. Pascon J.P., Coda H.B. Analysis of elastic functionally graded materials under
large displacements via high order tetrahedral elements. Finite elements in
analysis and design, 50: 33-47, 2012.
273. Phung-Van P., Nguyen-Thoi T. Luong-Van H., Lieu-Xuan Q. Geometrically
nonlinear analysis of functionally graded plates using a cell-based smoothed
three-node plate element (CS-MIN3) based on the C0-HSDT, Computer methods
in applied mechanics and engineering, 270: 15-36, 2014.
274. Pica A., Wood R.D. Post buckling behaviour of plates and shells using a Mindlin
Shallow shell formulation. Computers and Structures, 12: 759-768, 1980.
320
275. Pica A., Wood R.D., Hinton E. Finite element analysis of geometrically non
linear plate behavior using a Mindlin formulation, Computers and Structures, 11:
203-215, 1980.
276. Pietraszkiewicz W. On exact expressions of the bending tensor in the nonlinear
theory of thin shells. Applied Mathematical Modelling, 36:1821-1824, 2012.
277. Pietraszkiewicz W. Szymczak C. Shell Structures: Theory and Applications,
Taylor & Francis, London, 11-18, 2005.
278. Ping Zhu, Zhang L.W., Liew K.M. Geometrically non linear thermomechanical
analysis of moderately thick functionally graded plates using a Petro-Galerkin
approach with moving Kringing interpolation, Composite Structures, 107: 298-
314, 2014.
279. Pradhan S.C., Loy C.T., Lam K.Y., Reddy J.N. Vibration characteristics of
functionally graded cylindrical shells under various boundary conditions. Journal
of Applied Acoustics, 61:111-129, 2000.
280. Pradyumna S., Bandyopadhyay J.N. Free vibration analysis of functionally
graded curved panels using a higher-order finite element formulation. Journal of
Sound and Vibration, 318: 176-192, 2008.
281. Prakash T., Sundararajan N., Ganapathi M. On the nonlinear axisymmetric
dynamic buckling behavior of clamped functionally graded spherical cap, Journal
of Sound and Vibration, 299: 36-43, 2007.
282. Prathan G., Somashekar B.R. Field and edge consistency synthesis of a 4-noded
quadrilateral pla„ie bending element. International Journal of Numerical Methods
in Engineering, 26: 1692-1708, 1988.
283. Praveen G.N., Chin C.D., Reddy J.N. Thermoelastic analysis of functionally
graded ceramic-metal cylinder. Journal of Engineering Mechanics, 125(11):
1259-1267, 1999.
284. Praveen G.N., Reddy J.N. Non linear transient thermo elastic analysis of
functionally graded ceramic-plate plates. International Journal of Solids and
Structures, 35(33): 4457-4476, 1998.
285. Pulmano V.A., Lim P.K. Elastic analysis of skew thick plates by finite prism
method. International Conference on Computer Applications in Civil
Engineering, India. 121-126, 1979.
321
286. Put S., Vleugels, J., Van der Biest O. Microstructural engineering of
functionally graded materials by electrophoretic deposition. Journal of Material
Processing and Technology, 143-144: 572-577, 2003.
287. Qian L.F., Batra R.C., Chen L.M. Static and dynamic deformations of thick
functionally graded elastic plates by using higher-order shear and normal
deformable plate theory and meshless local Petrov–Galerkin method. Composites
Part B: Eng, 35:685-97, 2004.
288. Raju B. B., Shah C. G. Bending of swept cantilever thin elastic plates. National
Aeronautical Laboratory, Bangalore, TN-SA-7-66, 1966.
289. Ramesh C.K., Dixit V.D., Belkune R.M. Analysis of skew slab-beam system by
finite element method. Journal of Institution of Engineers India, 53: 233-239,
1973.
290. Reddy J.N. Analysis of functionally graded plates. International Journal of
Numerical Methods in Engineering, 47(1–3): 663-684, 2000.
291. Reddy J.N. A simple higher-order theory for laminated composite plates. Journal
of Applied Mechanics, 51(2):745–752, 1984b.
292. Reddy J.N. Analysis of layered composite plates accounting for large deflections
and transverse shear strains. In Recent Advances in Non linear computational
mechanics, Pineridge Press, Swansea, 1982.
293. Reddy J.N. Geometrically nonlinear transient analysis of laminated composite
plates. AIAA Journal, 21(4): 621-629, 1984a.
294. Reddy J.N. Mechanics of Laminated Composite Plates and Shells: Theory and
Analysis, second ed. CRC Press, Boca Raton, FL, 2004.
295. Reddy J.N. Nonlocal theories for bending, buckling and vibration of beams.
International Journal of Engineering Science, 45:288-307, 2007.
296. Reddy J.N. Thermo Mechanical Behavior of Functionally Graded Materials.
Texas, 1998.
297.Reddy J.N., Arciniega R. Shear deformation plate and shell theories: from
Stavsky to present. Mechanics of Advanced Materials and Structures, 11: 535-
582, 2004.
298. Reddy J.N., Chandrashekhara K. Geometrically non-linear transient analysis of
laminated, doubly curved shells. International Journal of Non-Linear Mechanics,
20: 79-90, 1985.
322
299. Reddy J.N., Jinseok Kim. A nonlinear modified couple stress-based third-order
theory of functionally graded plates. Composite Structures, 94: 1128-1143, 2012.
300. Reddy J.N., Khdeir A.A. Buckling and vibration of laminated composite plate
using various plate theories. AIAA Journal, 27(12):1808-1817, 1989.
301. Reddy J.N., Liu C.F. A higher-order shear deformation theory of laminated
elastic shells. International Journal of Engineering Sciences. 23: 319-330, 1985.
302. Reddy JN, Chin CD. Thermo mechanical analysis of functionally graded
cylinders and plates. Journal of Thermal Stresses, 21: 593-626, 1998.
303.Reissner E. On Bending of Elastic Plates. Quarterly of Applied Mechanics, 5:55-
68, 1947.
304. Reissner E. On the Theory of Bending of Elastic Plates. Journal of Mathematics
and Physics, 23(4):184-191, 1944.
305. Reissner E. Pure bending and twisting of skewed plates. Applied Mathematics,
10: 395-397, 1952.
306. Reissner E. Stein M. Torsion and transverse bending of cantilever plates.
N.A.C.A., Tech. Note 2369, 1951.
307. Reissner E. The effects of Transverse Shear Deformation on the Bending of
Elastic Plates. ASME Journal of Applied Mechanics, 12(9):A69-A77, 1945a.
308. Reissner, E. The effect of shear deformation on the bending of elastic plates.
Journal of Applied Mechanics, 12(2): 69-72, 1945b.
309. Reiter T., Dvorak G.J., Tvergaard V. Micromechanical models for graded
composite materials. Journal of Mechanics of Physics and Solids, 45: 1281-1302,
1997.
310. Ren J.G. A new theory of laminated plate. Composite Science and Technology,
26(3), 225-239, 1986.
311. Robert D.Cook, David S.Malkus, Michael E.Plesha and Robert J.Witt.
Concepts and applications of finite element analysis, John Wiley & Sons, New
York, 2004.
312. Rossow M. P. Observations on numerical modelling of an obtuse corner of a
simply supported plate. ASME Journal of Applied Mechanics, 45: 689-690, 1978.
313. Ruston K. R. Electrical analogue solutions for deformation of skew plates.
Aeronautics Q. XV: 169-180, 1964.
314. Salil Haldar. Free vibration of composite skewed cylindrical shell panel by finite
element method, Journal of Sound and Vibration, 311: 9-19, 2008.
323
315. Salvatore Brischetto. Classical and mixed advanced models for sandwich plates
embedding functionally graded cores. Journal of Mechanics of Materials and
Structures, 4(1):13-33, 2009.
316. Sanders L.J. Nonlinear theories for thin shells. Quarterly of Applied
Mathematics, 21: 21-36, 1963.
317. Sansour C. A theory and finite element formulation of shells at finite
deformations involving thickness change: circumventing the use of a rotation
tensor. Archive of Applied Mechanics, 65: 194-216, 1995.
318. Sarath Babu C., Kant T. Two shear deformable finite element models for
buckling analysis of skew fibre-reinforced composite and sandwich panels.
Composite Structures 46: 115-124, 1999.
319. Sattinger S. S., Conway H. D. Solution of certain isosceles triangle and rhombus
torsion and plate problems. International Journal of Mechanical Sciences 7: 221-
228, 1965.
320. Sengupta D. Performance study of a simple finite element in the analysis of skew
rhombic plates. Computers and Structures, 54: 1173-1182, 1995.
321. Serge Abrate. Free vibration, buckling and static deflections of functionally
graded plates. Composites Science and Technology, 66: 2383-2394, 2006.
322. Shankar C.A., Iyenger N.G.R. A C0 element for the free vibration analysis of
laminated composite plates. Journal of Sound and Vibration, 191(5): 721-738,
1996.
323. Sheikh A.H., Haldar S., Sengupta D. Vibration of plates in different situations
using a high-precision shear deformable element. Journal of Sound and Vibration,
253(2): 329-345, 2002.
324. Sheikh A.H., Mukhopadhyay M. Geometric nonlinear analysis of stiffened
plates by the spline finite strip method. Computers and Structures, 76: 765-785,
2000.
325. Shen G.G., Wang X. Thermal vibration, buckling and dynamic stability of
functionally graded cylindrical shells embedded in an elastic medium. Journal of
Reinforced Plastics and Composites, 27: 117-134, 2008.
326. Shen H.S. Nonlinear bending response of functionally graded plates subjected to
transverse loads and in thermal environments. International Journal of
Mechanical Sciences, 44:561–584, 2002.
324
327. Shen H.S. Nonlinear vibration of shear deformable FGM cylindrical shells
surrounded by an elastic medium. Computers and Structures, 94: 1144-1154,
2012.
328. Shimpi R .P. Refined plate theory and its variants. AIAA Journal, 40(1): 137-146,
2002.
329. Shimpi R.P., Patel H. G. Free vibrations of plate using two variable refined plate
theory. Journal of Sound Vibration, 296(4-5): 979-999, 2006a.
330. Shimpi R.P., Patel H.G. A two variable refined plate theory for orthotropic plate
analysis. International Journal of Solids Structure, 3(22): 6783-6799, 2006 b.
331. Shiyekar S.M., Kant T. An Electromechanical Higher Order Model for
Piezoelectric Functionally Graded Plates. International Journal of Mechanics and
Materials in Design, 6:163-174, 2010.
332. Shyang-Ho Chi, Yen-Ling Chung. Mechanical behavior of functionally graded
material plates under transverse load-Part I: Analysis. International Journal of
Solids and Structures, 43: 3657-3674, 2006.
333. Singh B.N., Iyengar N.G.R., Yadav D. A C0 finite element investigation for
buckling of shear deformable laminated composite plates with random material
properties. International Journal of Structural Engineering and Mechanics,
113(1): 53-74, 2002.
334. Singha M.K., Prakash T., Ganapathi M. Finite element analysis of
functionally graded plates under transverse loads. Finite elements in analysis and
design, 47:453-460, 2011.
335. Slimane Merdaci, Abdelouahed Tounsi, Mohammed Sid Ahmed Houari,
Ismail Mechab, Habib Hebali, Samir Benyoucef. Two new refined shear
displacement models for functionally graded sandwich plates. Archive of Applied
Mechanics, 81:1507-1522, 2011.
336. Soldatos K.P. A transverse shear deformation theory for homogeneous
monoclinic plates. Acta Mechanica, 94(3): 195-220, 1992.
337. Song Xiang, Gui-wen Kang, Ming-sui Yang, Yan Zhao. Natural frequencies
of sandwich plate with functionally graded face and homogeneous core.
Composite Structures, 96, 226-231, 2013.
338. Srinivas S., Joga Rao C.V., Rao A.K. An exact analysis for vibration of simply
supported homogeneous and laminated thick rectangular plates. Journal of Sound
and Vibration, 12:187-99, 1970.
325
339. Srinivas S., Rao A.K. Bending, vibration and buckling of simply supported
thick orthotropic plates and laminates. International Journal of Solids and
Structures, 6:1463-1481, 1970.
340. Srinivasan R.S., Bobby W. Non linear analysis of skew plates using the finite
element method. Computers and Structures, 6: 199-202, 1976.
341. Stavsky V. Pure bending, twisting and stretching of skewed heterogeneous
anisotropic plates. AIAA Journal, 1: 221-222, 1963.
342. Steigmann D. Koiter's shell theory from the perspective of three dimensional
nonlinear elasticity. Journal of Elasticity, 111: 91–107, 2013.
343. Suresh S., Mortensen A. Fundamentals of functionally graded materials. 1st ed.
London: IOM Communications, 1998.
344. Swaminathan K. Patil S.S. Analytical solutions using a higher order refined
computational model with 12 degrees of freedom for the free vibration analysis of
antisymmetric angle-ply plates. Composite Structures, 82:209-216, 2008.
345. Talha M, Singh B.N. Thermo-mechanical induced vibration characteristics of
shear deformable functionally graded ceramic-metal plates using the finite
element method, Proceedings of the Institution of mechanical engineers, Part C:
Journal of Mechanical Engineering Science, 225(50): 50-65 , 2011.
346. Talha M., Singh B.N. Static response and free vibration analysis of FGM plates
using higher order shear deformation theory. Applied Mathematical modeling,
34(12): 3991-4011, 2010.
347. Talha M., Singh B.N. Thermo-mechanical buckling analysis of finite element
modelled functionally graded ceramic-metal plates. International Journal of
Applied Mechanics, 3(4): 867-880, 2011.
348. Tani J., Liu G.R. Surface waves in functionally gradient piezoelectric plates.
JSME International Journal Series A (Mechanics and Materials Engineering)
36(2):152-155, 1993.
349. Todd A. Anderson. A 3-D elasticity solution for a sandwich composite with
functionally graded core subjected to transverse loading by a rigid sphere.
Composite Structures, 60: 265-274, 2003.
350. Toorani M.H., Lakis A.A. General equations of anisotropic plates and shells
including transverse shear deformations, rotary inertia and initial curvature
effects. Journal of Sound and Vibration 237:561-615, 2000.
326
351. Tornabene F. 2-D GDQ solution for free vibrations of anisotropic doubly-curved
shells and panels of revolution. Composite Structures, 93:1854-1876, 2011.
352. Tornabene F. Vibration analysis of functionally graded conical, cylindrical and
annular shell structures with a four-parameter power-law distribution. Computer
Methods in Applied Mechanics and Engineering, 198: 2911-2935, 2009b.
353. Tornabene F., Viola E., Inman D.J. 2-D differential quadrature solution for
vibration analysis of functionally graded conical, cylindrical and annular shell
structures. Journal of Sound and Vibration, 328: 259-229, 2009a.
354. Touratier M. An efficient standard plate theory. International Journal of
Engineering Science, 29(8):901-916, 1991.
355. Tsung-Lin Wu, K.K. Shukla, Jin H. Huang. Post-buckling analysis of
functionally graded rectangular plates. Composite Structures, 81: 1-10, 2007.
356. Tsung-lin Wu, Shukla K.K., Jin. H. Huang. Nonlinear static and dynamic
analysis of functionally graded plates. International Journal of Applied
Mechanics and Engineering, 11(3), 679-698, 2006.
357. Upadhyay A.K., Shukla K.K. Geometrically non linear static and dynamic
analysis of functionally graded skew plates. Communications in Nonlinear
Sciences and Numerical Simulations, 18: 2252-2279, 2013a.
358. Upadhyay A.K., Shukla K.K. Post-buckling behavior of composite and
sandwich skew plates. International Journal of Non-Linear Mechanics, 55:120-
127, 2013b.
359. Vaghefi R., Baradaran G.H., Koohkan H. Three-dimensional static analysis of
thick functionally graded plates by using meshless local Petrov-Galerkin
(MLPG) method. Enginering analysis with boundary elements, 34: 564-573,
2010.
360. Vanmeensel K., Anne G., Jiang D., Vleugels J., Van der Biest O. Processing of
a graded cutting tool in the Al203-Zr02-Ti(C,N) system by electrophoretic
deposition. Material Science Forum, 492-493: 705-710, 2005.
361. Vel S.S., Batra R.C. Three-dimensional exact solution for the vibration of
functionally graded rectangular plates. Journal of Sound and Vibration, 272:703-
30, 2004.
362. Venugopal Rao D., Sheikh A.H., Mukhopadhyay M. A finite element large
displacement analysis of stiffened plates. Computers and Structures, 47(6): 987-
993, 1993.
327
363. Victor Birman, Larry W.Byrd. Modelling and analysis of functionally graded
materials and structures. Applied Mechanics Reviews, 60: 195-216, 2007.
364. Vora M. R., Matlock W. Anisotropic skew plates and grids. ASCE Journal of
Engineering 105: 237-253, 1979.
365. Vosoughi A.R., Malekzadeh P., Banana Mo.R., Banan Ma.R. Thermal post
buckling of laminated composite skew plates with temperature-dependent
properties. Thin-Walled Structures, 49: 913-922, 2011.
366. Wang D. W. Katz I. N., Sxabo B. A. h and p-version of finite element analysis
of a rhombic plate. .International Journal of Numerical Methods in Engineering
20: 1399-1405, 1984.
367. Wang S. Buckling analysis of skew fibre-reinfored composite laminates based on
first order shear deformation theory. Composite Structures, 37: 5-19, 1997.
368.Wang X., Tan M., Zhou Y. Buckling analyses of anisotropic plates and isotropic
skew plates by the new version differential quadrature method. Thin-Walled
Structures, 41: 15-29, 2003.
369. Wankhade R. L. Geometric non linear analysis of skew plates using finite
element method. International Journal of Advanced Engineering Technology,
2(2): 154-163, 2011.
370. Warren W. E. Bending of rhombic plates. AIAA Journal l2: 166-168, 1964.
371. Watanabe Y., Nakamura Y., Fukui Y., Nakanishi K. A magnetic-functionally
graded material manufactured with deformation induced martensitic
transformation. Journal of Material Science, 12(5):326-328, 1993.
372. Whitney J.M., Leissa A.W. Analysis of Heterogeneous Anisotropic Plates.
ASME Journal of Applied Mechanics, 36:261-266, 1969.
373. Whitney J.M., Pagano N.J. Shear Deformation in Heterogeneous Anisotropic
Plates. ASME Journal of Applied Mechanics, 37:1031-1036, 1970.
374. Woo J., Meguid S.A. Non linear analysis of functionally graded plates and
shallow shells. International Journal of Solids and Structures, 38: 7409-7421,
2001.
375. Woo J., Meguid S.A., Stranart J.C., Liew K.M. Thermo mechanical post
buckling analysis of moderately thick functionally graded plates and shallow
shells. International Journal of Mechanical Sciences, 47: 1147-1171, 2005.
328
376. Woodward B., Kashtalyan M. 3D elasticity analysis of sandwich panels with
graded core under distributed and concentrated loadings. International Journal of
Mechanical Sciences, 53: 872-885, 2011.
377. Xiang S., Wang K.M, Ai Y.T., Sha Y.D. Thin plate spline radial basis functions
for vibration analysis of clamped laminated composite plates. European Journal
of Mechanics A/Solids, 29: 844-850, 2010.
378. Xiang S., Wang K.M., Ai Y.T., Sha Y.D., Shi H. Analysis of isotropic,
sandwich and laminated plates by a meshless method and various shear
deformation theories. Composite Structures, 91(1), 31-37, 2009.
379. Xiao J.R., Batra R.C., Gilhooley D.F., Gillespie J.W., McCarthy M.A.
Analysis of thick plates by using a higher-order shear and normal deformable
plate theory and MLPG method with radial basis functions. Computer Methods in
Applied Mechanics and Engineering, 196 (4-6): 979-987, 2007.
380. Xiao-Lin Huang, Hui-Shen Shen. Non-linear vibration and dynamic response of
functionally graded plates in thermal environments. International Journal of
Solids and Structures, 41: 2403-2427, 2004.
381. Xiaoqin S. Kaitai L. Yang M. The modified model of Koiter's type for
nonlinearly elastic shell. Applied Mathematical Modelling, 34: 3527-3535, 2010.
382. Xue Wang, Hui Fan, Whye-Teong Ang. On micromechanical-statistical
modeling of microscopically damaged interfaces under antiplane deformations.
International Journal of Solids and Structures, DOI:
ttp://dx.doi.org/10.1016/j.ijsolstr.2014.03.001.
383. Yadav D., Kamle S., Talukdar S. Transverse vibratin of multi mass loaded
variable section beams under random excitation. Journal of Sound and Vibration,
207(2):151-173, 1997.
384. Yamaoka H, Yuki M, Tahara K, Irisawa T, Watanabe R, Kawasaki A.
Fabrication of functionally gradient material by slurry stacking and sintering
process. Ceramic Transactions, Functionally Gradient Materials 1993; 34:165-
72.
385. Yang J., Liew K.M., Wu Y.F., Kitipornchai S. Thermo-mechanical post-
buckling of FGM cylindrical panels with temperature-dependent properties.
International Journal of Solids and Structures, 43: 307-324, 2006.
329
386. Yang J., Shen H.S. Vibration characteristics and transient response of shear-
deformable functionally graded plates in thermal environments. Journal of Sound
and Vibration, 255(3): 579-602, 2002.
387. Yang J., Shen., Hui-Shen. Free vibration and parametric resonance of shear
deformable functionally graded cylindrical panels. Journal of Sound and
Vibration, 261(5): 871-893, 2003.
388. Yang P.C., Norris C.H., Stavsky Y. Elastic Wave Propagation in
Heterogeneous Plates. International Journal of Solids and Structures, 2(4):665-
684, 1966.
389. Yang Z., Chen X., Li B., He1 Z., Miao H. Vibration analysis of curved shell
using B-Spline wavelet on the interval (BSWI) finite elements method and
general shell theory. CMES 85: 129-155, 2012.
390. Yanga J., Hui-Shen Shen. Non-linear analysis of functionally graded plates
under transverse and in-plane loads. International Journal of Non-Linear
Mechanics, 38: 467-482, 2003.
391. Yaqoob Yasin M., Kapuria S. An efficient layerwise finite element for shallow
composite and sandwich shells. Composite Structures, 98: 202-214, 2013.
392. Yettram A. L. An equivalent grid framework for skew plates in flexure.
International Journal of Mechanical Sciences, 14: 407-416, 1972.
393. Zenkert D. The handbook of sandwich construction. London: Engineering
Materials Advisory Services Ltd. (EMAS), 1997.
394. Zenkour A.M. A comprehensive analysis of functionally graded sandwich
plates: Part 1-Deflection and stresses. International Journal of Solids and
Structures, 42(18–19):5224-5242, 2005 a.
395. Zenkour A.M. A comprehensive analysis of functionally graded sandwich
plates: Part 2-Buckling and free vibration. International Journal of Solids and
Structures, 42(18–19): 5243-5258, 2005b.
396. Zenkour A.M. Analytical solution for bending of cross-ply laminated plates
under thermo-mechanical loading. Composite Structures, 65: 367-379, 2004a.
397. Zenkour A.M. Buckling of fiber-reinforced viscoelastic composite plates using
various plate theories. Journal of Engineering Mathematics, 50: 75-93, 2004b.
398. Zenkour A.M. Generalized shear deformation theory for bending analysis of
functionally graded plates. Applied Mathemtical Modelling, 30(1): 67-84, 2006.
330
399. Zenkour A.M. Thermal effects on the bending response of fiber-reinforced
viscoelastic composite plates using a sinusoidal shear deformation theory. Acta
Mechanica, 171: 171-187, 2004 c.
400. Zenkour A.M., Sobhy M. Thermal buckling of various types of FGM sandwich
plates. Composite Structures, 93:93-102, 2010.
401. Zhang D.G., Zhou Y.H. A theoretical analysis of FGM thin plates based on
physical neutral surface, Computational Material Science, 44: 716-720, 2008.
402. Zhang L.W., Zhu P., Liew K.M. Thermal buckling of functionally graded plates
using a local Kriging meshless method. Composite Structures, 108: 472-492,
2014.
403. Zhao X., Lee Y.Y., Liew K.M. Free vibration analysis of functionally graded
plates using the element-free kp-Ritz method. Journal of Sound and Vibration,
319(3-5): 918-939, 2009a.
404. Zhao X., Lee Y.Y., Liew K.M. Mechanical and thermal buckling analysis of
functionally graded plates. Composite structures, 90: 161-171, 2009b.
405. Zhao X., Lee Y.Y., Liew K.M. Thermo elastic and vibration analysis of
functionally graded cylindrical shells. International Journal of Mechanical
sciences, 51: 694-707, 2009c.
406. Zhao X., Liew K.M. A mesh-free method for analysis of the thermal and
mechanical buckling of functionally graded cylindrical shell panels.
Computational Mechanics 45:297-310, 2009a.
407. Zhao X., Liew K.M. An element-free analysis of mechanical and thermal
buckling of functionally graded conical shell panels. International Journal of
Numerical Methods in Engineering 86: 269-285, 2011.
408. Zhao X., Liew K.M. Geometrically non linear analysis of functionally graded
plates using the element-free kp-Ritz method. Computer Methods in Applied
Mechanics and Engineering, 198: 2796-2811, 2009b.
409. Zhao X., Liew K.M. Geometrically non linear analysis of functionally graded
shells. International Journal of Mechanical Sciences, 51: 131-144, 2009c.
410. Zhou D., Cheung Y.K., Au F.T.K., Lo S.H. Three-dimensional vibration
analysis of thick rectangular plates using Chebyshev polynomials and Ritz
method. International Journal of Solids and Structures, 39: 6339-6353, 2002.
411. Zhou L., Zheng W.X. Vibration of skew plates by the MLS-Ritz method.
International Journal of Mechanical Sciences, 50: 1133-1141, 2008.
331
412. Zhu J.C., Yin Z.D., Lai Z.H. Fabrication and microstructure of ZrO2-Ni
functional gradient material by powder metallurgy. Journal of Materials Science,
31: 5829-5834, 1996.
413. Zhu P., Liew K.M. Free vibration analysis of moderately thick functionally
graded plates by local Kriging meshless method. Composite Structures 93:2925-
44, 2011.
414. Zienkiewicz 0. C., Lefebvre D. A robust triangular plate bending element of the
Reissner-Mindlin type. International Journal of Numerical Methods in
Engineering, 26: 1169-1184, 1988.
332
APPENDIX A
A.1 Transformation matrix for constant variation of transverse
displacement field
c s 0 0 0 0 0
 
s c 0 0 0 0 0
0 0 1 0 0 0 0
 
[T ]   0 0 0 c s 0 0  , where c  cos and s  sin
0 0 0 s c 0 0
 
0 0 0 0 0 c s
0 0 c 
 0 0 0 s
A.2 Transformation matrix for quadratic variation of transverse

displacement field
 c s 0 0 0 0 0 0 0 0 0 0 0
 
s c 0 0 0 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0 0 0 0 0
 
0 0 0 c s 0 0 0 0 0 0 0 0
0 0 0 s c 0 0 0 0 0 0 0 0
 
0 0 0 0 0 c s 0 0 0 0 0 0
T    0 0 0 0 0 s c 0 0 0 0 0 0  , where c  cos and s  sin

0 0 0 0 0 0 0 c s 0 0 0 0
0 0 0 0 0 0 0 s c 0 0 0 0 

0 0 0 0 0 0 0 0 0 c s 0 0
 
0 0 0 0 0 0 0 0 0 s c 0 0
0 0 0 0 0 0 0 0 0 0 0 c s 
 
0 0 0 0 0 0 0 0 0 0 0 s c 
333
APPENDIX B
B.1 [H]p1 matrix for FGM plate considering constant variation of

 4 z 3 
1 0 0 0 0 z 0 0 0 0 0 0 0 0 
 3h 2 
 4 z 3 
0 1 0 0 0 0 z 0 0 0 0 0 0 0 
 3h 2 
 4 z 3 
 H p  0
3h 2 
1
0 1 0 0 0 0 z 0 0 0 0 0 0

 4 z 2 
0 0 0 1 0 0 0 0 1 0
h2
0 0 0 0 
 
 4 z 2 
0 0 0 0 1 0 0 0 0 1 0
h2
0 0 0 

B.2 [H]p2 matrix for FGM plate considering quadratic variation of

1 0 0 0 0 0 z 0 0 0 0 0 z2 0 0 0 0 z3 0 0
 
0 1 0 0 0 0 0 z 0 0 0 0 0 z2 0 0 0 0 z3 0
0 0 1 0 0 0 0 0 z 0 0 0 0 0 0 0 0 0 0 0
 H p  
2
0 0 0 1 0 0 0 0 0 z 0 0 0 0 z2 0 0 0 0 0
0 0 0 0 1 0 0 0 0 0 z 0 0 0 0 z2 0 0 0 0
 
0 0 0 0 0 1 0 0 0 0 0 z 0 0 0 0 z2 0 0 z 3 
B.3 [H]G matrix for FGM plate considering constant variation of

0 0 0 0 0 0 0 0 1 0 0 0 0 0 
0 0 0 0 0 0 0 0 0 1 0 0 0 0 

 4 z 3 
1 0 0 0 z 0 0 0 0 0 0 0 0 
 3h 2 
 4 z 3 
H G   0 0 1 0 0 0 z 0 0 0 0 0 0 
 3h 2 
 3
0 4 z 
0 0 1 0 0 0 z 0 0 0 0 0
 3h 2 
 
 4 z 3 
0 1 0 0 0 z 0 0 0 0 0
3h 2
0 0 

334
B.4 [H]G matrix for FGM plate considering quadratic variation of
1 0 0 0 0 0 z 0 0 0 0 0 z2 0 0 0 0 0 0 0 0 0
 
0 1 0 0 0 0 0 z 0 0 0 0 0 z2 0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 z 0 0 0 0 0 z2 0 0 0 z3 0 0 0
 HG    
0 0 0 1 0 0 0 0 0 z 0 0 0 0 0 z2 0 0 0 z3 0 0
0 0 0 0 1 0 0 0 0 0 z 0 0 0 0 0 z2 0 0 0 z3 0
 
0 0 0 0 0 1 0 0 0 0 0 z 0 0 0 0 0 z2 0 0 0 z 3 
335
APPENDIX C
C.1 Non-linear strain terms
 u   v   w 
2 2 2
 nlxx   0    0    0  ;
 x   x   x 
0
 u0  x   v0  y  w0  z 

knl1  2 
xx
.   2 .   2 . ;
 x x   x x   x x 
   x   v0  y
   y 
2
 u   w0  z    z 
2 2
k xx
  0 . x   
 .   2 .   ;
 x x   x   x x
   
nl2
  x   x x   x 
8  u   8  u       8  v  y 
knlxx3   2  0 . x  2  0 . x  x . x  2  0 . 
3h  x x  3h  x x   x x  3h  x x 
8  v0  y    y  y  2  u0  x  2  v0  y    z  z 
  .  .   .   .   2 . ;
3h 2  x x   x x  3  x x  3  x x   x x 
8   y  8   y  y 
2
8    8     1   x 
2 2
k xx
 2 x   2 x. x      2    2 
. 
3h  x  3h  x x  4  x  3h  x  3h  x x 
nl4
1   y  2   x  x  2   y  y
2
   z 
2
     .   .   ;
4  x  3  x x  3  x x   x 
8   x  x  8   x  x  8   y  y  8   y  y 
knlxx5   .  2  .  2  .  2  . 
6h 2  x x  6h  x x  6h  x x  6h  x x 
1    x  1   y  y 
  x.   . ;
3  x x  3  x x 
16   y  32   y  y 
2
16    32     16   
2 2
k xx
 4  x   4  x . x  4  x   4    4 . 
9h  x  9h  x x  9h  x  9h  x  9h  x x 
nl6
16   y  8   y  y 
2
8   x  x 8   x  x  z   x 
2

6
 4   2 .  2  .      2 . 
9h  x  9h  x x  9h  x x  9  x  9h  x x 
8   y  y  z 6   y 
2
 2 .    ;
9h  x x  9  x 
2 2 2
 u   v   w 
 yy
 0   0   0  ;
 y   y   y 
nl0
 u    v  y   w0  z 
knyyl1  2 0 . x   2 0 .   2 . ;
 y y   y y   y y 
   x   v0  y    y 
2 2 2
 u   w0  z    z 
k yy
  0 . x    .    2 .   ;
 y y   y   y 
nl2
  y   y   y   y   y 
336
 8  u0  x    x  x  8  v0  y 
8  u0  x
knlyy3    2 
 . .  .  2  . 
3h 2  y y
 3h  y y   y y  3h  y y 
8  v  y    y  y  2  u0  x  2  v0  y    z  z 
 2 0.  .   .   .   2 . ;
3h  y y   y y  3  y y  3  y y   y y 
8   y  8   y  y 
2 2 2
8    8     1   
k yy
 2  x   2  x . x   x   2    2 . 
3h  y  3h  y y  4  y  3h  y  3h  y y 
nl4
1   y  2   x  x  2   y  y
2 2
   z 
     .  
  ;.
4  y  3  y y  3  y y
  y 
8     8     8   y  y  8   y  y 
knlyy5  2  x . x  2  x . x  2  .  2  . 
6h  y y  6h  y y  6h  y y  6h  y y 
1     1   y  y 
  x. x   . ;
3  y y  3  y y 
16   y  32   y  y 
2 2 2
16    32     16   x 
k yy
 4 x  4 x. x   4       4 
. 
9h  y  9h  y y  9h  y  9h  y  9h  y y 
nl6 4
16   y  8   y  y 
2 2
8   x  x  8   x  x  z 6   x 
 4   2 .  2  .     2 . 
9h  y  9h  y y  9h  y y  9  y  9h  y y 
8   y  y  z 6   y 
2
 2 .    ;
9h  y y  9  y 
 nlzz   x    y    z  ;
2 2 2
0
knlzz1  2  x . x   2  y . y   4 z ;
  2  x . x    x   2  y   2  y . y    y   4 z2  2 x . x
8 8 8 8
2  x
2 2
knlzz2  
2 2
h h h h
 2 y . y ;
 .  2  x . x   2  y . y   2  y . y   2 x . x  2 y . y ;
8 8 8 8
2  x x
knlzz3 
3h 3h h 3h
 4  x   4  x . x   4   x   4  y   4  y . y   4   y 
16 2 32 16 2 16 2 32 16 2
knlzz4
h h h h h h
 2  x . x    x   2   x . x   2  y . y    y   2   y . y  ;
8 2 8 8 2 8
h h h h
 u   v   w 
 nlyz   x . 0    y . 0     z . 0  ;
0
 y   y   y 
 u       z   v0    y   w0 
knlyz1     x . 0     x . x    z.    y .    y .   2z . ;
 y   y   y   y   y   y 
337
4  u0  4  v0  4  u0  4  v0    x 
knlyz2   .
2  x   2  y .   2  x.   2  y.    x. 
h  y  h  y  h  y  h  y   y 
  y  1   x  1   y   u0   v0    z    z 
  y .    x .    y .    x .    y .    z.   2z. ;
 y  2 y  2 y   y   y   y   y 
4    4   y  4   x  4   y  4   x 
knlyz3   2   x . x   2   y .   
 x .   
 y .   . x  
h  y  h  y  h 2  y  h 2  y  3h 2  y 
4   y  1   x  1   y  4   x  4   y    y 
 
.     .     .  2  . x   2  . y    . y 
3h  y y  2  y  3h  y  3h  y   y
2 y x y
 2 
1   y    x  1      
  y .  . x     x . x   2   z . z  ;
3 y   y  3 y   y 
4    4   y  2   x  2   y 
knlyz4  2   x . x   2   y .    x .    y . 
3h  y  3h  y  h 2  y  h 2  y 
2    2   y  4   x  4   y 
 2  x. x   2   y.   2  x .   2  y . 
h  y  h  y  3h  y  3h  y 
1  y  1  y  1   x  1   x 
  y .    y .    x .    x . ;
2 y  3  y  2  y  3  y 
16    16   y  16   x  16   y  16   x 
knlyz5  4  x . x   4  y .   4  x.   4  y.   4  x . 
3h  y  3h  y  3h  y  3h  y  3h  y 
16   y  16   x  16   y  4   x  4   x 
  y .   4  x.   4  y.   2  x .   2  x . 
3h 4  y  3h  y  3h  y  3h  y  3h  y 
4    4    1    4   y  4   y 
 2   x . x   2   x . x    x . x   2  y .   2  y . 
3h  y  3h  y  3  y  3h  y  3h  y 
4   y  4   y  1   y 
 2  y .   2  y.    y . ;
3h  y  3h  y  3  y 
 u   v   w 
 nlxz0    x . 0     y . 0     z . 0  ;
 x   x   x 
 u         v    y   w0 
knlxz1     x . 0     x . x     z . z     y . 0     y .   2z . ;
 x   x   x   x   x   x 
4  u  4  v  4  u  4  v       y 
knlxz2   2   x . 0   2   y . 0   2   x . 0   2   y . 0     x . x     y . 
h  x  h  x  h  x  h  x   x   x 
1    1   y   u0   v0    z    z 
  x . x    y .    x .    y .    z .   2z. ;
2 x  2  y   x   x   x   x 
338
4   x  4   y
 4   x  4   y  4   x 
knlxz3    x .   2  y .
  2  x.   2  y.   . x 
h2  x  h   h  x x  h  x  3h 2  x 
4   y  1   x  1   y  4   x  4   y    y 
 2 . y     x .     .  2  . x   2  . y    . y 
3h  x x  2  x  3h  x  3h  x   x
y
 2 
1   y    x  1      
  y .  . x     x . x   2   z . z  ;
3 x   x  3 x   x 
4    4   y  2   x  2   y  2   x 
knlxz4  2  x . x   2  y .    x .    y .   2  x. 
3h  x  3h  x  h 2  x  h 2  x  h  x 
2  y  4   x  4   y  1   y  1   y 
 2  y.   2  x .   2  y .    y .    y . 
h  x  3h  x  3h  x  2  x  3  x 
1   1   
  x . x     x . x  ;
2 x  3  x 
16    16   y  16   x  16   y  16   x 
knlxz5  4   x . x   4   y .   4  x.   4  y.   x . 
3h  x  3h  x  3h  x  3h  x  3h 4  x 
16   y  16   x  16 
 4   y  x  4   x 
  y .   4  x.   4  y.
  2  x .   2  x . 
3h 4  x  3h  x  3h 
 3h  x x  3h  x 
4    4    1    4   y  4   y 
 2   x . x   2   x . x    x . x   2  y .   2  y . 
3h  x  3h  x  3  x  3h  x  3h  x 
4   y  4   y  1   y 
  y .   2  y.    y . ;
3h 2  x 3h  x  3  x 
 u u   v v   w w 
 nlxy   0 . 0    0 . 0    0 . 0 ;
 x y   x y   x y 
0
 u    u0  x   v0  y
  v0  y   w0  z   w0  z 
knlxy1   0 . x  .  
. .  .  . ;
 x y   y x   x y
  y x   x y   y x 
1  u       1   u  1  v  y    y  y  1   y v0 
knlxy2    0 . x  x . x   x . 0   0 .  .   . 
2  x y   x y  2  x y  2  x y   x y  2  x y 
     w0  z   w0  z 
 z . z  .  . ;
 x y   x y   y x 
4  u0  x
 4  u0  x  1   x  x  4  u0  x  1   x  x 
knlxy3    . 2  .   .  2  .   . 
3h 2  x y
 3h  x y  2  y x  3h  y x  2  x y 
4   u  4  v  y  4  v0  y  1   y  y  4  v0  y 
 2  x . 0  2  0 .   .   .   . 
3h  x y  3h  x y  3h 2  x y  2  y x  3h 2  y x 
339
1   y  y  4   y v0  1  v0  y  1  v0  y  1  u0  x 
  .  2  .   .   .   . 
2  x y  3h  x y  3  x y  3  y x  3  x y 
1  u     z  z    z  z 
  0. x  .  . ;
3  y x   x y   y x 
8   x  x
 4   x  x  1   x  x  4   x  x  8   y  y 
knlxy4    .  2  .   .  2  .  2  . 
3h 2  x y 3h  x y  4  x y  3h  x y  3h  x y 
4   y  y  1   y  y  4   y  y  1   x  x  1   y  y 
 2 .   .   .   .   . 
3h  x y  4  x y  3h 2  x y  3  x y  3  y x 
1     1   y  y    z  z 
  x . x   .  . ;
3  y x  3  x y   y x 
4   x  x  4   x  x  4   x  x  4   x  x  4   y  y 
knlxy5   .   .   .   .   . 
6h 2  x y  6h 2  y x  6h 2  x y  6h 2  y x  6h 2  x y 
4   y  y  4   y  y  4   y  y  1   x  x  1   y  y 
 2 .   .   .   .   . 
6h  x y  6h 2  y x  6h 2  x y  6  x y  6  x y 
1     1   y  y 
  x . x   . ;
6  y x  6  y x 
16     16     16     16     16   y  y 
knlxy6  4  x . x   4  x . x   4  x . x   4  x . x  4  . 
9h  x y  9h  x y  9h  x y  9h  x y  9h  x y 
16   y  y  16   y  y  16   y  y  8   x  x  8   x  x 
 4 .   .   .   .   . 
9h  x y  9h 4  x y  9h 4  x y  9h 2  x y  9h 2  x y 
 8   x  x  1   x  x  8   y  y
8   x  x  8   y  y 
  .  2  .   .  2  .  2  . 
9h 2  y x 9h  y x  9  y x  9h  x y  9 h  x y 
8   y  y  8   y  y  1   y  y 
 2 .   .   . ;
9h  y x  9h 2  y x  9  y x 
340
APPENDIX D
D.1 Different terms of matrix [A]
u0 v w
A1,1  ; A1,3  0 ; A1,5  0 .
x x x
u v w
A2,2  0 ; A2,4  0 ; A2,6  0 .
y y y
A3,7   x ; A3,8   y ; A3,9   z .
2u0 2v0 2w0
A4,7  ; A4,8  ; A4,9  .
y y y
2u0 2v0 2w0
A5,7  ; A5,8  ; A5,9  .
x x x
2u0 2v0 2w0
A6,2  ; A6,4  ; A6,6  .
x x x
2u0 2v0 2w0
A7 ,10  ; A7,12  ; A7,14  .
x x x
2u0 2v0 2w0
A8,11  ; A8,13  ; A8,15  .
y y y
A9,16  2 x ; A9,17  2 y ; A9,18  4 z .
 x  y  u v w
A10,7  2 ; A10,8  2 ; A10,9  2 z ; A10,16  2 0 ; A10,17  2 0 ; A10,18  4 0 .
y y y y y y
  y  u v w
A11,7  2 x ; A11,8  2 ; A11,9  2 z ; A11,16  2 0 ; A11,17  2 0 ; A11,18  4 0 .
x x x x x x
u u v v w w
A12,10  2 0 ; A12,11  2 0 ; A12,12  2 0 ; A12,13  2 0 ; A12,14  2 0 ; A12,15  2 0 .
y x y x y x
  y  u v w
A13,10  x ; A13,12  ; A13,14  z ; A13,19   0 ; A13,21   0 ; A13,23  2 0 .
x x x x x x
  y  u v w
A14,11  x ; A14,13  ; A14,15  z ; A14,20   0 ; A14,22   0 ; A14,24  2 0 .
y y y y y y
8 8 8
A15,7   2  x ; A15,8   2  y ; A15,16   x ; A15,17   y ; A15,18  4 z ; A15,25   2  x ;
h h h
8
A15,26   2  y ; A15,31  2 x ; A15,32  2 y .
h
8 u 8 v    y 
A16,7   2 0 ; A16,8   2 0 ; A16,9  2 z ; A16,16  2 x ; A16,17  2 ; A16,18  4 z ;
h y h y y y y y
341
8 u0 8 v u v
A16,20   x ; A16,22   y ; A16,25   ; A16,26   2 0 ; A16,31  2 0 ; A16,32  2 0 .
h y
2
h y y y
8 u 8 v    y 
A17,7   2 0 ; A17,8   2 0 ; A17,9  2 z ; A17,16  2 x ; A17,17  2 ; A17,18  4 z ;
h x h x x x x x
8 u 8 v u v
A17,20   x ; A17,22   y ; A17 ,25   2 0 ; A17,26   2 0 ; A17,31  2 0 ; A17,32  2 0 .
h x h x x x
  y  u u v
A18,11  2 x ; A18,13  2 ; A18,14  2 z ; A18,19   0 ; A18,20   0 ; A18,21   0 ;
x x x y x y
v w w
A18,22   0 ; A18,23  2 0 ; A18,24  2 0 .
x y x
8 u 8 v   y 
A19,10   2 0 ; A19,12   2 0 ; A19,19   x ; A19,21   ; A19,23  2 z ;
3h x 3h x x x x
8 u 8 v 2 u0 2 v0
A19,27   2 0 ; A19,28   2 0 ; A19,33   ; A19,35   .
3h x 3h x 3 x 3 x
8 u 8 v   y 
A20,11   2 0 ; A20,13   2 0 ; A20,20   x ; A20,22   ; A20,24  2 z ;
3h y 3h y y y y
8 u 8 v 2 u0 2 v0
A20,28   2 0 ; A20,29   2 0 ; A20,34   ; A20,36   .
3h y 3h y 3 y 3 y
8 8 8 8
A21,16  2  x ; A21,17  2  y ; A21,25  2  x ; A21,26  2  y ; A21,31  2 x ; A21,32  2 y .
h h h h
32 32
A22,11   2  x ; A22,13    2  y ; A22,20   x ; A22,22   y ; A22,24  4 z ;
3h 3h
8  8  y 8 8
A22,25   2 x ; A22,26   2 ; A22,28   2  x ; A22,30   2  y ;
h y h y 3h 3h
  y 2 2
A22,31  2 x ; A22,32  2 ; A22,34    x ; A22,36    y .
y y 3 3
32 32
A23,10   2  x ; A23,12   2  y ; A23,19   x ; A23,21   y ; A23,23  4 z ;
3h 3h
8  8  y 8 8
A23,25   2 x ; A23,26   2 ; A23,27   2  x ; A23,29   2  y ;
h x h x 3h 3h
  y 2 2
A23,31  2 x ; A23,32  2 ; A23,33    x ; A23,35    y .
x x 3 3
342
8  x 8  8  y 8  y
A24,1   ; A24,2   2 x ; A24,3   2 ; A24,4   2 ;
3h y
2
3h x 3h y 3h x
   y  y 
A24,19   x ; A24,20   x ; A24,21   ; A24,22   ; A24,23  2 z ;
y x y x y
 8 u 8 u 8 v
A24,24  2 z ; A24,27   2 0 ; A24,28   2 0 ; A24,29   2 0 ;
x 3h y 3h x 3h y
8 v 2 u0 2 u0 2 v0
A24,30   2 0 ; A24,33   ; A24,34   ; A24,35   ;
3h x 3 y 3 x 3 y
2 v0
A24,36   .
3 x
8  8  y 1  x 1  y 
A25,10   2 x ; A25,12   2 ; A25,19  ; A25,21  ; A25,23  z ;
3h x 3h x 4 x 4 x x
8  8  y 2  x 2  y
A25,27   2 x ; A25,29   2 ; A25,33   ; A25,35   .
3h x 3h x 3 x 3 x
8  8  y 1  x 1  y 
A26,11   2 x ; A26,13   2 ; A26,20  ; A26,22  ; A26,24  z ;
3h y 3h y 4 y 4 y y
8  8  y 2  x 2  y
A26,28   2 x ; A26,30   2 ; A26,34   ; A26,36   .
3h y 3h y 3 y 3 y
 16 8   16 8   16 32 
A27,7   4  x  2  x  ; A27,8   4  y  2  y  ; A27,25   4  x  4  x  ;
h h  h h  h h 
 16 32   8   8 
A27,26   4  y  4  y  ; A27,31   x  2  x  ; A27,32   y  2  y  .
h h   h   h 
8  8  y 4 4 4  x
A28,16  2 x ; A28,17  2 ; A28,20  2  x ; A28,22  2  y ; A28,25  2 ;
3h y 3h y h h h y
4  y 8 8  x  y
A28,26  2 ; A28,28  2  x ; A28,30  2  y ; A28,31  ; A28,32  ;
h y 3h 3h y y
2 2
A28,34   x ; A28,36   y .
3 3
8  8  y 4 4 4  x
A29,16  2 x ; A29,17  2 ; A29,19  2  x ; A29,21  2  y ; A29,25  2 ;
3h x 3h x h h h x
4  y 8 8  x  y
A29,26  2 ; A29,27  2  x ; A29,29  2  y ; A29,31  ; A29,32  ;
h x 3h 3h x x
343
2 2
A29,33   x ; A29,35   y .
3 3
16  16  y 1  x 1  y
A30,11   2 x ; A30,12   2 ; A30,20  ; A30,21  ;
3h x 3h y 2 x 2 y
 8  8  8  y 8  y
A30,24  2 z ; A30,27   2 x ; A30,28   2 x ; A30,29   2 ; A30,30   2 ;
x 3h y 3h x 3h y 3h x
2  x 2  x 2  y 2  y
A30,33   ; A30,34   ; A30,35   ; A30,36   .
3 y 3 x 3 y 3 x
4  4  y 4  x 4  y
A31,19  2 x ; A31,21  2 ; A31,27  2 ; A31,29  2 ;
3h x 3h x 3h x 3h x
1  x 1  y
A31,33  ; A31,35  .
3 x 3 x
4  4  y 4  x 4  y
A32,20  2 x ; A32,22  2 ; A32,28  2 ; A32,30  2 ;
3h y 3h y 3h y 3h y
1  x 1  y
A32,34  ; A32,36  .
3 y 3 y
 32  32  x   32  y 32  y   32  x 32  x 
A33,7   4 x  4  33,8  4
; A   4  ; A33,25   4  4 ;
 3h y 3h y   3h y 3h y   3h y 3h y 
 32  y 32  y  8 8  8  x 2  x 
A33,26   4  4  ; A33,28  2  x ; A33,30  2  y ; A33,31   2  ;
 3h y 3h y  3h 3h  3h y 3 y 
 8  y 2  y   8 8   8 8 
A33,32   2   ; A33,34   2  x  2  x  ; A33,36   2  y  2  y  .
 3h y 3 y   3h 3h   3h 3h 
 32  32  x   32  y 32  y   32  x 32  x 
A34,7   4 x  4  34,8  4
; A   4  ; A34,25   4  4 ;
 3h x 3h x   3h x 3h x   3h x 3h x 
 32  y 32  y  8 8  8  x 2  x 
A34,26  4  4  ; A34,27  2  x ; A34,29  2  y ; A34,31   2  ;
 3h x 3h x  3h 3h  3h x 3 x 
 8  y 2  y   8 8   8 8 
A34,32  2   ; A34,33   2  x  2  x  ; A34,35   2  y  2  y  ;
 3h x 3 x   3h 3h   3h 3h 
4  x 4  x 4  y 4  y
A35,19  ; A  ; A  ; A  ;
3h 2 y 3h 2 x 3h 2 y 3h 2 x
35,20 35,21 35,22
4  x 4  x 4  y 4  y
A35,27  2 ; A35,28  2 ; A35,29  2 ; A35,30  2 ;
3h y 3h x 3h y 3h x
344
1  x 1  x 1  y 1  y
A35,33  ; A35,34  ; A35,35  ; A35,36  .
3 y 3 x 3 y 3 x
 16  32    16  y 32  y   16  x 8  x 
A36,10   4 x  4 x  ; A36,12   4  4  ; A36,27   4  2 ;
 9h x 9h x   9 h  x 3 h  x   9 h  x 9 h  x 
 16  y 8  y   8  x 1  x   8  y 1  y 
A36,29   4  2 
 36,33  2
; A   36,35  2
; A   .
 9 h x 9 h x   9 h  x 9 x   9 h x 9 x 
 16  32    16  y 32  y   16  x 8  x 
A37,11   4 x  4 x  ; A37,13   4  4  ; A37,28   4  2 ;
 9h y 9h y   9 h y 3h y   9 h y 9 h  y 
 16  y 8  y   8  x 1  x   8  y 1  y 
A37,30   4  2 
 37,34  2
; A  
 37,36  2
; A  .
 9 h  y 9 h y   9 h y 9 y   9 h  y 9 y 
 32  32   32   32  y 32  y 
A38,10   4 x  4 x  ; A38,11  4 x ; A38,12   4  4 ;
 9h y 9h y  9h x  9 h y 9 h y 
32  y  32  8  x   32  y 8  y 
A38,13  4 ; A38,27   4 x  2 
 38,29  4
; A  ;
9h x  9h y 9h y   9 h  y 9 h 2
y 
 8  1  x   8  x 2  x 
A38,33   2 x   ; A38,34   2  ;
 9h y 9 y   9h x 9 x 
 8  y 1  y   8  y 2  y 
A38,35   2   38,36  2
; A   .
 9h y 9 y   9h x 9 x 
345
LIST OF PUBLICATIONS
INTERNATIONAL JOURNALS
1. Gulshan Taj M.N.A., Anupam Chakrabarti, Abdul Hamid Sheikh. Analysis of

functionally graded plates using higher order shear deformation theory. Applied
Mathematical Modelling, 37: 8484-8494, 2013.
2. Gulshan Taj M.N.A., Anupam Chakrabarti. Static and dynamic analysis of
functionally graded skew plates. ASCE Journal of Engineering Mechanics, 139(7):
848-857, 2013.
3. Gulshan Taj M.N.A., Anupam Chakrabarti. Buckling analysis of functionally
graded skew plates: an efficient C0 finite element approach, International Journal of
Applied Mechanics, 5(4): 1350041-1350060, 2013.
4. Gulshan Taj M.N.A., Anupam Chakrabarti. An efficient Co finite element approach
for bending analysis of functionally graded ceramic-metal skew shell panels. Journal
of Solid Mechanics, 5(1): 47-62, 2013.
5. Gulshan Taj M.N.A., Anupam Chakrabarti. Dynamic response of functionally
graded skew shell panel. Latin American Journal of Solids and Structures, 10: 1243-
1266, 2013.
6. Gulshan Taj M.N.A., Anupam Chakrabarti. Bending analysis of functionally graded
skew sandwich plates with through-the thickness displacement variations. Journal of
Sandwich structures and Materials, 16(2): 210-248, 2014.
7. Gulshan Taj M.N.A., Anupam Chakrabarti, Vipul Prakash. Vibration characteristics
of functionally graded material skew plate in thermal environment. International
Journal of Mechanical and Aerospace Engineering (Accepted).
8. Gulshan Taj M.N.A., Anupam Chakrabarti. Free vibration analysis of four
parameter functionally graded plate accounting for realistic transverse shear mode,
Vietnam Journal of Mechanics (Accepted).
9. Gulshan Taj M.N.A., Anupam Chakrabarti. Modeling of functionally graded
sandwich shells accounting for variation in transverse displacement. Journal of
Mechanics of Advanced Materials and Structures (Communicated).
346
10. Gulshan Taj M.N.A., Anupam Chakrabarti. On the modeling of four-variable
power law functionally graded plates using an efficient and accurate theory.
Proceedings of the Institution of Mechanical Engineers: Part C (Communicated).
11. Gulshan Taj M.N.A., Anupam Chakrabarti. Stability study on four-parameter
power function based functionally graded plate models under thermo-mechanical
environment. Chinese Journal of Aeronautics (Communicated).
12. Gulshan Taj M.N.A., Anupam Chakrabarti. Geometrically nonlinear analysis of
shear deformable functionally graded ceramic-metal sandwich plates with symmetric
layups. Finite Elements in Analysis and Design (Communicated).
13. Gulshan Taj M.N.A., Anupam Chakrabarti. Large deformation analysis of spherical
and hyperbolic paraboloid sandwich shells incorporating homogeneous/FGM core.
Computer methods in Applied mechanics and Engineering (Communicated).
14. Gulshan Taj M.N.A., Anupam Chakrabarti. Large deformation response of
functionally graded cylindrical, hyperbolic paraboloid and elliptic paraboloid panels
accounting for realistic transverse deformation. Archive of Applied Mechanics
(Communicated).
15. Gulshan Taj M.N.A., Anupam Chakrabarti. Nonlinear bending response of
functionally graded spherical and hypar panels via higher order finite element model.
ASCE Journal of Engineering Mechanics (Communicated).
16. Gulshan Taj M.N.A., Anupam Chakrabarti. 2D FE model for post buckling analysis
of functionally graded sandwich plates with symmetric layups. Composite Structures
(Communicated).
17. Gulshan Taj M.N.A., Anupam Chakrabarti. Non-linear 2D modelling of shear

deformable functionally graded symmetric cylindrical and hypar sandwich shells.
International Journal of Solids and Structures (Communicated).
INTERNATIONAL CONFERENCES
1. Gulshan Taj M.N.A., Anupam Chakrabarti. Analysis of functionally graded skew

plate under mechanical loads, 3rd Asian Conference on Mechanics of Functional
Materials and Structures (ACMFMS), Indian Institute of Technology Delhi, India,
December 5-8, 2012.
347
2. Gulshan Taj M.N.A., Anupam Chakrabarti. Bending analysis of functionally
graded sandwich skew plates, International Conference on Structural Engineering
and Mechanics (ICSEM), National Institute of Technology, Rourkela, December 20-
22, 2013.
3. Gulshan Taj M.N.A., Anupam Chakrabarti. C0 finite element formulation for static
analysis of functionally graded ceramic - metal skew shells, 1st International
conference on Mechanics of Composites, Stony Brook University, USA, June 8-12,
2014 (Accepted).
4. Gulshan Taj M.N.A., Anupam Chakrabarti. Non linear bending study of shear
deformable functionally graded plates using efficient finite element method, The
Second Australasia and South-East Asia Structural Engineering and Construction
Conference Bangkok, Thailand, November 3-7, 2014 (Accepted).
348

Final Thesis-Print Out Taken-2!4!14

Uploaded by

Copyright:

Available Formats

You might also like

Final Thesis-Print Out Taken-2!4!14

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Final Thesis-Print Out Taken-2!4!14

Uploaded by

Copyright:

Available Formats

FINITE ELEMENT ANALYSIS OF FUNCTIONALLY GRADED

PLATES AND SHELLS

DEPARTMENT OF CIVIL ENGINEERING

DEPARTMENT OF CIVIL ENGINEERING

(Gulshan Taj M.N.A)

The Ph.D. Viva-Voce Examination of Ms. Gulshan Taj M.N.A, Research

Supervisor Chairman, SRC External Examiner

Head of the Department/Chairman,ODC

Functionally graded materials (FGMs) are advanced class of engineering composites

I would like to take this opportunity to acknowledge the members of my examining

Finally, I am thankful to the authorities of Indian Institute of Technology, Roorkee for

Indian Institute of Technology (Gulshan Taj M.N.A)

List of Figures xvi-xxiii

List of Tables xxiv-xxx

CHAPTER 1 INTRODUCTION 1-10

1.2 Objectives and scope of the present research 9

1.3 Organization of the thesis 10

CHAPTER 2 REVIEW OF LITERATURE 11-80

2.2 Brief review of various shear deformation theories of

2.3 Static, dynamic and buckling responses of FGM plates/shells 21

2.4 Static, dynamic and buckling responses of skew plates/shells 38

2.5 Static, dynamic and buckling responses of sandwich

2.6 Geometrically non-linear responses of laminated and

2.7 Post buckling response of laminated and functionally graded

2.8 Appraisal from previous research works 77

3.2 Homogenization of FGMs 85

3.2.1 Self consistent scheme 85

3.2.2 Mori-Tanaka approach 86

3.2.3 Composite sphere assemblage model 86

3.2.4 Composite cylindrical assemblage model 87

3.2.5 The simplified strength of materials method 87

3.2.6 The method of cells 87

3.2.7 Micromechanical models 87

3.3 Idealization of FGMs 87

3.3.1 The Exponential law 88

3.3.2 The Power law 88

3.4 Geometrical Configuration 88

3.4.1 Single layer FGM plate/shell 88

3.4.2 Sandwich FGM plate/shell 92

3.4.2.1 Sandwich FGM plate/shell with FGM skin 93

3.4.2.2 Sandwich FGM plate/shell with FGM core 95

3.5 Four-parameter power law distribution 97

3.6 Skew boundary transformation 98

3.7 Thermal analysis 99

3.8 Assumptions 100

3.9 Material modeling 100

3.9.1 Constitutive relationship 100

3.9.2 Element description 101

3.10 Displacement field 103

3.10.1.2 Strain-displacement relationship for shell 105

3.10.2 Displacement field having quadratic variation of

3.10.2.2 Strain-displacement relationship for shell 110

3.10.3 Governing equation for static analysis 111

3.11 Governing equation for dynamic analysis 113

3.11.1 Free vibration analysis 113

3.11.2 Forced vibration analysis 114

3.12 Governing equation for buckling analysis 115

3.13 Geometrically non linear analysis 116