See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/245339867

Examination of indirect responses of helicopters using a refined inflow model

Article  in  Aircraft Engineering and Aerospace Technology · January 2009

DOI: 10.1108/00022660910926881

CITATIONS READS
3 77

2 authors:

Farid Shahmiri Fariborz Saghafi

8 PUBLICATIONS   7 CITATIONS   
Sharif University of Technology
35 PUBLICATIONS   154 CITATIONS   
SEE PROFILE
SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Tandem rotor UAV View project

All content following this page was uploaded by Farid Shahmiri on 09 May 2017.

The user has requested enhancement of the downloaded file.

 Aircraft Engineering and Aerospace Technology
Examination of indirect responses of helicopters using a refined inflow model
Farid Shahmiri Fariborz Saghafi
Article information:
To cite this document:
Farid Shahmiri Fariborz Saghafi, (2009),"Examination of indirect responses of helicopters using a refined inflow model", Aircraft
Engineering and Aerospace Technology, Vol. 81 Iss 1 pp. 25 - 37
Permanent link to this document:
http://dx.doi.org/10.1108/00022660910926881
Downloaded on: 25 December 2014, At: 20:55 (PT)
References: this document contains references to 41 other documents.
To copy this document: permissions@emeraldinsight.com
The fulltext of this document has been downloaded 711 times since 2009*
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

Access to this document was granted through an Emerald subscription provided by 203778 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service
information about how to choose which publication to write for and submission guidelines are available for all. Please visit
www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of
more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online
products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics
(COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.

*Related content and download information correct at time of download.

 Examination of indirect responses
of helicopters using a refined inflow model
Farid Shahmiri and Fariborz Saghafi
Aerospace Department, Sharif University of Technology, Tehran, Iran

Abstract
Purpose – The purpose of this paper is to examine the cross-coupled responses of a coupled rotor-fuselage flight dynamic simulation model, including
a finite-state inflow aerodynamics and a coupled flap-lag and torsion flexible blade structure.
Design/methodology/approach – The methodology is laid out based on model development for an articulated main rotor, using the theories of
aeroelastisity, finite element and finite-state inflow formulation. The finite-state inflow formulation is based on a 3D unsteady Euler-based concepts
presented in the time domain. The most advantages of the model are the capability of modeling dynamic wake effects, tip losses and skewed wake
aerodynamics. This is, in fact, a special type of the inflow model relating inflow states, to circulatory blade loadings through a set of first-order
differential equations. A non-iterative solution of the differential equations has practically altered the model into a simple and direct formulation
appending properly to the rest of the helicopter mathematical model. A non-linear distribution of the induced velocity over the rotor disc is finally
obtained by the use of both Legendre polynomials and higher-harmonic functions. Ultimately, validations of the theoretical results show that the on-axis
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

response, direct reaction to the pilot input, has a good accuracy both quantitatively and qualitatively against flight test data, and the off-axis response,
cross-coupled or indirect reaction to the pilot input are improved by this approach of modeling.
Findings – Improvements in dynamic prediction of both trim control settings and dynamic cross-coupled responses of helicopter to pilot inputs are
observed.
Research limitations/implications – Further work is required for investigation of the augmented finite state inflow model, including the wake
rotation correction factors to describe helicopter maneuvering flight characteristics.
Practical implications – The results of this work support the future researches on design and development of advanced flight control system,
incorporating a high bandwidth with low-phase delay to control inputs and also high levels of dynamic stability within minimal controls cross coupling.
Originality/value – This paper provides detailed characteristics on the mathematical integration problems associated with the advanced helicopter
flight dynamics research.

Keywords Simulation, Helicopters, Flow, Flight dynamics

Paper type Research paper

Nomenclature b0 ¼ blade coning angle

x ¼ wake skew angle
CT ¼ steady-state thrust coefficient
V ¼ main rotor speed
e ¼ hinge offset (n)!! ¼ double factorial of n; nðn 2 2Þ. . .ðn 2 4Þ 2
Lo, Lno ¼ circulatory and non-circulatory blade lift for n even; nðn 2 2Þ. . .ðn 2 4Þ 1 for n odd
Li ¼ non-dimensional circulatory lift of the ith (. . .) * ¼ derivative with respect to azimuth angle
blade ( ¼ Li/rV2R3) (. . .)ˆ ¼ dimensionless quantity
Nb ¼ number of blades (. . .)T ¼ transpose index
ub , v b , w b ¼ elastic displacements (. . .)I ¼ inertial reference frame ¼ (1I, 2I, 3I)
vij ¼ angular velocity of jth relative to ith (. . .)B ¼ body-fixed reference frame ¼ (1B, 2B, 3B)
coordinate system (. . .)S ¼ shaft-fixed reference frame ¼ (1S, 2S, 3S)
x0, y0, z0 ¼ location of a point on the blade local (. . .)U ¼ un-deformed blade reference frame
coordinate system ¼ (1U, 2U, 3U)
m ¼ helicopter advanced ratio ( ¼ V/RV) (. . .)K ¼ deformed blade reference frame
mX, mY, mZ ¼ advanced ratio components in the tip path ¼ (1K, 2K, 3K)
plane frame (. . .)L ¼ local blade sectional reference frame
fb ¼ elastic twist ¼ (1L, 2L, 3L)
cb ¼ blade azimuth angle
s ¼ rotor solidity, stress components

The current issue and full text archive of this journal is available at
www.emeraldinsight.com/1748-8842.htm
Aircraft Engineering and Aerospace Technology: An International Journal
81/1 (2009) 25– 37
q Emerald Group Publishing Limited [ISSN 1748-8842]
[DOI 10.1108/00022660910926881]
1. Introduction
Generally, helicopter flight dynamics simulation is the task of
analyzing the dynamic response of a helicopter as a whole,
particularly, helicopter response during unsteady transient
conditions. The flight dynamics analysis inherently involves
three fundamental sections. The first section is to describe the
dynamics of the various components of the helicopter in an

25
 Indirect responses of helicopters using a refined inflow model
Farid Shahmiri and Fariborz Saghafi
appropriate mathematical form. The second element involves
the coupling of each component to form a complete model
describing the dynamics of the helicopter as a whole. The
third section is to use this complete model for determination
of trim equilibrium conditions and time history responses to
arbitrary pilot control inputs.
Although some significant efforts performed around for a
number of years have been slightly promoted the dynamic
responses of helicopters, the reports show a long-term
occurrence described off-axis problem as a major disruption
in explanation of the final results (Celi, 2005; Lovera et al.,
2007; Ribera and Celi, 2004, 2005, 2007a,b; Shahmiri and
Saghafi, 2007; Theodore and Celi, 2003; Tischler, 1999;
Fletcher and Tischler, 1997; Simons and Modha, 2002). This
probably means that the most of the former studies are
somewhat sophisticated when compared to real flights.
Since most of the pervious investigations are commonly
focused on the aerodynamics mishaps as a main source of the off-
axis discrepancy, the examination of advanced main rotor inflow
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)
aerodynamics modeling compatible to flight dynamics
simulation requirements can be an effective style to improve
the problem. For this reason, in the current paper, some aspects
of improving are presented through the application of the refined
inflow modeling. From the flight dynamic points of view, each of
the inflow models has specific strengths and weaknesses in
explanation of the flight dynamic responses. Many of the simple
induced inflow models that have been around are still in widely
use today in studying different types of flight dynamics problems.
The reasons are that these models are computationally
inexpensive, in the form required to be attached to the existing
codes formulated in first-order form, and they can be easily
embedded into the formulation and solution techniques. Some
comprehensive reviews of inflow modeling and its use in flight
dynamics analyses have been carried out by Gaonkar and Peters
(1986, 1988) and Chen (1989). In these reviews, the inflow
models are limited to a set of simple inflow models. Pitt and
Peters (1981) is the most widely used of these models presented a
linear distribution of inflow over the rotor disc by utilizing three
first-order differential equations. The main challenge of using
this model is that it limits the inflow to a linear radial variation
with only the first harmonic azimuthally variation over the rotor
disk. The second point is the low-frequency nature of this inflow
model, and its inability to accurately work in the high-frequency
cases. The kind of results that are of our interest would include
higher-harmonic blade loading and high-frequency loads on the
helicopter (Peters et al., 1989; He, 1989; He et al., 1999;
von Grünhagen, 1995; Basset, 1998). The next level of inflow
models is provided based on a prescribed vortex wake
(Basset, 1998 Basset and Elomari, 1999; Keller, 1996; Keller
and Curtiss, 1996, 1998; Arnold et al., 1995, 1998; Schulein,
1999; Rosen, 2004a, b; Rosen and Isser, 1995). Using the Biot-
Savart law, the induced velocity at any point on the rotor disk is
calculated based on the total influence of all the modeled vortices
associated with the wake.
There are a number of difficulties associated with these inflow
models. The first is that the wake geometry is prescribed. In
other words, it requires enough information about the
appropriate wake structure for a specific flight condition and
the helicopter configuration. A second difficulty is that the
inflow distribution is calculated based on a wake structures at a
specific instant of time, and thus, the time variation of the inflow
velocity and blade loading are not considered. The third
drawback is that the inflow cannot in general be represented by a
26
Aircraft Engineering and Aerospace Technology: An International Journal
Volume 81 · Number 1 · 2009 · 25 –37
set of first order equations. Hence, it cannot be simply
appended to the flight dynamics models as a dynamic inflow
model. This creates problems for various aspects of flight
dynamics analyses including the calculation of linearized
models generating numerically, because of no states
associated with the wake geometry or inflow distribution that
can be perturbed independent of the rest of the model.
The other approach is to use free wake models to generate the
rotor inflow distributions used in some flight dynamics
programs (Bagai et al., 1998; Park and Leishman, 1999;
Bhagwat and Leishman, 2001; Bhagwat, 2001; Theodore and
Celi, 2003). With these models, the rotor wake geometry is not
prescribed and is allowed distorting under the influence of the
flight condition, blade loading and self-induced interaction.
Although the free wake inflow models seems to have the least
amount of assumptions of all types, and seem the most realistic
wake structure among the inflow models, there are a number of
drawbacks accompanied by these models.
Some drawbacks of these models can be appeared through
the following questions:
.
How the rotor wake is to be modeled?
.
How the structure of the vortex filaments is to be
modeled?
. What is the initial strength of the vortex filaments, and
how the dissipation and diffusion of the vortices can be
modeled?
.
Where is the radial release point of the vortices from the
blades, and how the roll up of the vorticity in the tip
region is handled?
.
What are the characteristics of the discretization of the
vortex wake, and the age of the individual vortex filaments
before they are assumed to no longer influence the main
rotor inflow?
All of the previous studies have attempted to improve the off-
axis response prediction through refined theoretical models.
This indicates that each theoretical model only captures some
but not all the effects required to give the best agreement with
test data. It is fairly certain that the main source of the above-
mentioned off-axis coupling problem is rooted in rotor
aerodynamics. Therefore, the coupling effect can be captured
by a more precise attention to the aerodynamics principles.
In the present work, the more sophisticated representation
of the main rotor dynamics and aerodynamics leading to
improve both the on-axis and off-axis response is described.
This is a cause of merging the coupled flap-lag and torsion
elastic blade model along with an unsteady finite state inflow
model in this research. The finite-state inflow theory improves
the former Pitt-Peters inflow model by entering a higher-order
representation of inflow variation along the rotor radius and
higher-harmonic variations around the azimuth. The type of
results that are of our interest is the inclusion of higher-
harmonic blade loadings and higher-frequency loads on the
helicopter during hover and forward flight phase.
2. Some remarks on induced inflow model
The wake model coupled with the flight dynamics code is the
finite state wake model of Peters and He (1991). The basic
governing equations of this model are derived from linearized
Euler equations. The linearization is performed based on the
fact that the induced velocities are small perturbations
relative to free stream inflow. In the non-dimensional form,
 Indirect responses of helicopters using a refined inflow model Aircraft Engineering and Aerospace Technology: An International Journal
Farid Shahmiri and Fariborz Saghafi Volume 81 · Number 1 · 2009 · 25 –37

the linearized Euler equations yield the acceleration potential The number of inflow states arj , brj depends on the number of
term, the non-dimensional form of pressure distribution. harmonics r and the number of polynomials per harmonics j
Since the acceleration potential function can be chosen by the user. Additionally, the radial location x^ and the
mathematically written in the form of Laplace equation, it azimuth location c b of a point on the blade elastic axis
can be obtained by using linear superposition based on the required for the calculation are determined through the
unsteady and velocity gradient definitions. This means that presented discussion in the next section.
the acceleration potential function is originated from the The inflow states are obtained through the use of the
influence of the unsteady rate of velocities and the gradient of following differential equations defined in the tip path plane
velocities in the flow field. The separation of variables coordinate system as:
practically used here is the way to transform the non-
dimensional linearized Euler equations into a set of first order  n r o* 0 h rmc i21 n r o  mc 
4
p Hm
n aj þv Ljn aj ¼ tn
ODEs based on linear operators. Figure 1 shows a general
layout of the finite state wake model formation.  n r o* 0 h rms i21 n r o  ms  ð4Þ
While the normal induced velocity is the most significant of all
4
p Hm
n bj þv Ljn bj ¼ tn
other components, the reduced ODEs corresponded to the third
component of inflow vector is adequate for inflow calculations.
The solution of the normal inflow ODEs is obtained when In equation (4), the Sine and Cosine harmonic coupling
the forcing function together with normal induced velocity li matrices ½Lrms rmc
jn , ½Ljn  relating the jth polynomial of rth

is approximated by a set of radial modal functions cjr harmonic of the induced inflow due to nth polynomial of mth
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

(Legendre polynomial) and azimuth-wise harmonics as: harmonic of perturbation pressure field are obtained based on
the definition of the undistorted helical wake structure as
X
1 X
1 h follows:
li ð^x; c b ; ^tÞ ¼ cjr ð^xÞ arj ð^tÞcosðr c b Þ
r¼0 j¼rþ1; rþ3;... ð1Þ
i
þbrj ð^tÞsinðr c b Þ ð21Þðnþj22rÞ=2
Grm
jn ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Hm n Hj
r

where: pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 ð2n þ 1Þð2j þ 1Þ
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
r þ m ¼ even
X
n21
ð j þ nÞð j þ n þ 2Þ½ð j 2 nÞ2 2 1
cmn ð^xÞ ¼ ð2n þ 1ÞH m n x^ q ð21Þq2m=2 p ð5Þ
q¼m;mþ2;...
ð2Þ Grm
jn ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðm þ qÞ!! 2 Hm n Hj
r

ðq 2 mÞ!!ðq þ mÞ!!ðm 2 q 2 1Þ!! sgnðr 2 mÞ

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi r þ m ¼ odd j ¼ n ^ 1
ðn þ m 2 1Þ!!ðn 2 m 2 1Þ!! ð2n þ 1Þð2j þ 1Þ
Hm ¼ ð3Þ
n
ðn þ mÞ!!ðn 2 mÞ!! Grm
jn ¼ 0 r þ m ¼ odd j – n ^ 1

Figure 1 Schematic of finite state wake formation

Unsteady Euler Approximated

(PDEs Eqs.) Solution, Eq.(1)
Normal Induced
Velocity li
Linearized
Form
Normal Inflow
(First Order ODEs)
Matrix Form
Non-Dimensional Eq.(4)
Form

First Order ODEs

Induced Inflow
Acceleration
States ajr, bjr
Potential Concept
(Laplace Eqs.)

Solution
(Linear
Superposition)

27
 Indirect responses of helicopters using a refined inflow model Aircraft Engineering and Aerospace Technology: An International Journal
Farid Shahmiri and Fariborz Saghafi Volume 81 · Number 1 · 2009 · 25 –37

and: other types of lift models. In other words, the properties

involved in the finite state structure are limited factors in
h i selecting of this appropriate blade lift model having the least
L0mc
jn ¼ ðX m ÞG0m
jn amount of interference effects. The related expressions of this
h i h i
model, including lift, pitching moment and drag in the local
rmc
Ljn ¼ ðX jm2rj
þ ð21Þl X jmþrj Þ Grm
jn l ¼ minðr; mÞ
h i h i ð6Þ blade coordinate system, Figure 2, are finally corresponded to:
Lrms
jn ¼ ðX jm2rj 2 ð21Þl X jmþrj Þ Grm
jn 

x 1 2 € 1
  p L ¼ Lqs þ arðbRÞ h þ v0 a_b 2 xac 2 bR a€b
X ¼ tan   0 # x # 2 2
2 2   ð11Þ
_h a_b
Furthermore, the pressure coefficients and are tnmc tnms Lqs ¼ arbRv20 ab þ þ ðbR 2 xac Þ
v0 v0
corresponded to the lift distribution through the following
equations:

1 1
M ¼ Lqs xac þ arðbRÞ2 xac 2 bR
2 2
Nb Z 1   
1 X 1
t0c
n ¼ 0
Li f m cn d^x
h€ 2 ðxac 2 bRÞa€b ð12Þ
2p i¼1 e 2
a
Nb Z 1  1
1X 2 arv0 a_b ðbR 2 xac ÞðbRÞ2 2 rðbRÞ4 a€b
tnmc ¼ Li f m cm c Þd^ ð7Þ
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

n cosðm i x 2 16
p i¼1 e
Z  1
1X Nb 1
D¼ rcv20 Cd ðaY ; MÞ ð13Þ
tnms ¼ Li f m cm
n sinðm ci Þd^x 2
p i¼1 e
where a is the lift-curve slope, r is the air density, b is the non-
where: dimensional semi-chord length, R is the blade radius, ab is the
pitch angle of the blade section, h; h; _ h€ are the vertical
R c=2R movement, velocity and acceleration, respectively, v0 is the
2c=2R P y ðyÞexpðimy=^xÞdy
fm ¼ R c=2R ð8Þ free stream velocity, xac is the blade cross-sectional aerodynamic
2c=2R P y ðyÞdy center offset from the elastic axis.
The time rate of change of the total blade pitch angle and
It should be noted that the correction factor fm is essentially a the blade local angle of attack, Figure 2, are, respectively,
function of a chord-wise pressure distribution. Among the corresponded to:
various types of aerodynamics theory used for determination of
the chord-wise pressure distribution, the lifting line theory is the a_b ¼ u_0 2 u_1c sin cb þ u_1s cos cb þ f_b ð14Þ
proper one yielding the value of the correction factor equal to one.
This is a valuable simplification, because of the concentration of
bound vortices at y ¼ 0. Figure 2 Schematic of blade angle of attack
Since the energy is added to the flow passed through the
Main Rotor Hub
rotor, the free stream velocity, equation (4), must be corrected
through the momentum consideration as: 2L

mX þ ð2lm 2 mZ Þðlm 2 mZ Þ
v0 ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð9Þ Blade Root
v0 1L
m2X þ ðlm 2 mZ Þ2
vz φY βb vy
where:
CT vx
lm ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð10Þ 3L
2 mX þ ðlm 2 mZ Þ2
2

Blade Tip
It is apparent
pffiffiffi that the use of mean (steady) inflow expression
lm . 3a01 rather than the equation (10) actually changes a
set of differential equations, equation (4), to the non-linear 3K 2L Blade Chord
type of inflow model. This is a way used in this research to
capture more precisely the non-linear effects of flow during
αb
the inflow calculations.
LC φY αb
αY
3. Remarks on main rotor equations vz
v0
φY 2K
The blade lift model embedded to the flight dynamics code is
φb
two-dimensional quasi-steady model of Theodore and Celi
(2003). This is a kind of model that is more compatible with the 1L
vx
assumptions, shed wake, trailed wake and compressibility D
effects, associated formerly with the finite state model than the Elastic Center

28
 Indirect responses of helicopters using a refined inflow model Aircraft Engineering and Aerospace Technology: An International Journal
Farid Shahmiri and Fariborz Saghafi Volume 81 · Number 1 · 2009 · 25 –37

aY ¼ tan 21 ðvX tan ab þ vZ Þcos bb

ð15Þ ðvT ÞL ¼ ½vX vY vZ T ð19Þ
ðvX 2 vZ tan ab Þcos2 bb
where:
where: qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
vX v0 ¼ v2X þ v2Y þ v2Z ð20Þ
cos bb ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi ð16Þ
v2X þ v2Y
and:
As it can be seen, the equations (11)-(13) are functions of the 0 1
oncoming free stream flow, v0. Although there are many DvX 2 LvZ cos bb
approaches to find the free stream flow, here, the total velocity 1 B C
B DvY 2 LðvY vZ cos bb Þ=ðvX Þ C ð21Þ
ðF A ÞL ¼ ½ f 1 f 2 f 3 T ¼
and its components are calculated for a point located on the v0 @ A
blade elastic axis, Figure 3, through the special following LðvX Þ=ðcos bb Þ þ DvZ
operators:
vT ¼ P I r 2 þ ðP B r 2 þ w IB £ r 2 Þ þ ðP U r 3 þ w IU £ r 3 Þ And the pitching moment is corresponded to:
ð17Þ 0 1
þ ðP U r 4 þ w IU £ r 4 Þ þ ½00li T 0
B C
B 0 C
with:
B C
› B 1 2 C
B rðbRÞ v 2
C ð a ; MÞ C
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

Pi ; i ¼ Coordinate System ð18Þ ðM A ÞL ¼ B 2 0 m Y C ð22Þ

›t i B L C
B qs C
Bþ xac ð f 2 cos ab þ f 1 sin ab Þ C
B L C
where ri is the position vector, wij is the angular velocity vector @ A
of j relative to i coordinate system and li is the normal 2 18 arðbRÞ2 v0 a_b ðbR 2 xac Þ
induced velocity discussed in the pervious section. The total
velocity vector is necessarily determined by the evaluation of The blade section lift, drag and pitching moment coefficients
each term of the equation (17) in the un-deformed coordinate Cl, Cd, Cm are obtained from look-up tables as a function of
system through the application of the appropriate the local angle of attack and Mach number. Since, the main
transformation matrices (Shahmiri and Saghafi, 2007). rotor equations of motion are written in the un-deformed
Furthermore, because the blade lift model, equations blade coordinate system, the aerodynamic loads must be
(11)-(13), is formulated in the local blade coordinate system, transformed into the un-deformed coordinate system.
the total velocity vector, equation (17), must be transferred The inertia loads, main part of the main rotor equations,
into the local blade coordinate system. This leads to the are generally dependent on the absolute acceleration of a
following relations: point on the rotor blade. These inertia forces and moments

Figure 3 Schematic of helicopter coordinate systems

3K
Elastic Axis
Inertial Frame Deformed Blade 1U
1K
1I 2I r5 Un-deformed
s Blade Frame
2K
β0 1R
s
wb
vb
Hub-Rotating Frame
ψb
3R r4
3I 3U ub

x0
2U
r1 2R
Hub r3
2S
e

r2
1S

Body Fixed
CG
Frame 2B

1B t2 t1

3B 3S

29
 Indirect responses of helicopters using a refined inflow model Aircraft Engineering and Aerospace Technology: An International Journal
Farid Shahmiri and Fariborz Saghafi Volume 81 · Number 1 · 2009 · 25 –37

per unit span of a blade in the un-deformed blade coordinate

system are, respectively, as follows: 1 ›s ›s

1xx ¼ · 21
2 ›x ›x
ðð

1 ›s ›s
ðFI ÞU ¼ 2 rb P2I sdAb 1xy ¼ · ð30Þ
Ab
2 ›x ›y
ðð ð23Þ

 1 ›s ›s
ðMI ÞU ¼ 2 r b r5 £ P2I s dAb 1xz ¼ ·
Ab 2 ›x ›z

where:
where rb is the blade mass density, Ab is the blade section area
and P2I s is expressed as: s ¼ r1 þ r2 þ r3 þ r4 þ r5
0 1
ub þ e cos b0 þ x0 þ T 21 y0 þ T 31 z0
B C ð31Þ
P2I s ¼ P2I r 1 þ P2B r 2 þ w IB £ ðP B r 2 þ w IB £ r 2 Þ þ P B ðw IB £ r 2 Þ B vb þ T 22 y0 þ T 32 z0 C
¼B C
þ P2B r 3 þ w IU £ ðP U r 3 þ w IU £ r 3 Þ þ P U ðw IU £ r 3 Þ @ A
wb 2 e sin b0 þ T 23 y0 þ T 33 z0
þ P2U r 4 þ w IU £ ðP U r 4 þ w IU £ r 4 Þ þ P U ðw IU £ r 4 Þ
þ P2K r 5 þ w IK £ ðP K r 5 þ w IK £ r 5 Þ þ P K ðw IK £ r 5 Þ Tij is the elements of the transformation matrix from
ð24Þ deformed to un-deformed coordinate system described by
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

Shahmiri and Saghafi (2007). Since the main rotor equations

of motion ultimately presents in the un-deformed coordinate
where the operator P2i is defined as: system, once again the final expressions are transferred from

the deformed to the original un-deformed coordinate system.
›2
P2i ; ð25Þ
›t 2 i 4. Some remarks on solution techniques
The subsequent calculations of the main rotor inertial loads The solution procedure used in this study is based on the
are based on the evaluation of each term of the equation (24) finite-element Galerkin method. In other words, the non-
in the un-deformed coordinate system yielding the inertia linear partial differential equations of motion of the rotor
loads per unit span for a blade (Shahmiri and Saghafi, 2007). blades are transformed into the ODEs by using the finite
The blade structural loads calculations, the latest element Galerkin scheme. A finite element has the eleventh
components of the main rotor equations, derived in terms degrees of freedom contained of flap, lag displacement and
of elastic displacements ub, vb, wb and elastic twist fb of a rotation for only inboard and outboard nodes and torsional
point on the blade elastic axis are based on the Shamie and rotation for each of the three nodes of the element. Figure 4
Friedmann (1977) model. This is a type of model based on shows a finite element and its associated nodal degrees of
the Bernoulli-Euler beam assuming low to moderate freedom. Here, the vector of nodal degrees of freedom of a
deflections. The structural load expressions presented in the blade partitioned into the lag, flap and torsion part is
deformed blade coordinate system are corresponded to: corresponded to the vector with 5n þ 6 elements where n is
the number of blade finite elements:
ðð h iT
›v1 ›vnþ1 ›w1
ðF S ÞK ¼ tdAb ¼ f S1 f S2 f S3 ð26Þ Y n ¼ v1 . . .vnþ1 w1 . . .w nþ1
Ab ›x ›x ›x

T ð32Þ
ðð h iT ›wnþ1

f1 ; f2 ; . . .f2nþ1
ðM S ÞK ¼ ðr 5 £ tÞdAb ¼ mS1 mS 2 mS 3 ð27Þ ›x
Ab
The number of blade degrees of freedom resulting from the
where: finite element discretization is reduced by the use of modal
h iT
coordinate transformation:
t ¼ sxx txy txz ð28Þ
Y n ¼ ½Wy ð33Þ
The blade structural loads based on the linear elastic and The modal matrix [W] contained of the blade mode shapes in
isotropic material are: equation (33) is obtained through the calculation of blade
0 1 2 30 1
sxx E 0 0 1xx mass [M] and stiffness [K] matrices estimated by. using the
B C 6 7B C numerical procedure known as a finite difference method.
B txy C ¼ 6 0 2G 0 7B 1xy C ð29Þ
@ A 4 5@ A The natural frequencies and corresponded mode shapes
txz 0 0 2G 1xz required for construction of the modal matrix columns are
evaluated through the numerical solution of the free-vibration
problem as:
where the quantities E and G, the young and shear modulus
of the material, are taken from look-up tables. The € n þ ½KY n ¼ 0
½MY ð34Þ
components of the strain for a point on the blade are
obtained from the base vector of deformed section The sufficiently reduced number of degrees of freedom
›s=›x; ›s=›y; ›s=›z as: composed of the first four torsional, lag, and flap bending

30
 Indirect responses of helicopters using a refined inflow model Aircraft Engineering and Aerospace Technology: An International Journal
Farid Shahmiri and Fariborz Saghafi Volume 81 · Number 1 · 2009 · 25 –37

Figure 4 Definition of blade element degrees of freedom

Vb2 Wb2

Wb2,x
Finite Element

Wb1 Vb2,x f2
Vb1

Vb1,x
f1/2

Wb1,x
f1

Blade

Hinge
HUB
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

normal modes of a rotating blade is considered in a 2 3T

compromise between accuracy and computational expenses.
Since the rotor blades equations of motion, 12 ODEs
presented in terms of modal coefficients, are formulated in the
rotating coordinate system, the solution is also obtained in the
same coordinate system.
The first step of the solution process is the calculation of
trim states of helicopter, when the flight condition parameters
such as air-speed V, turn rate c_ and climb angle g for a
straight level and hover flight is specified. The unknown states
of trim problem obtained through the implementation of the
non-linear solver acting on a set of algebraic equations are
corresponded to:
2 3T
0 m
u0 u1c u1s u0t a ua01 . . . am n b1 . . . bn lt
x¼4 1 1 1 1 1
y0 y1c y1s y2c y2s . . . y1mc y1ms . . . yk0 yk1c yk1s yk2c yk2s . . .ykmc ykms
ð35Þ
where the steady state periodic motion of the blades in flap,
lag and torsion is provided in the form of truncated Fourier
series for the modal coefficients:
m 
0 m
u v w p q r f u c a01 . . . am
n b1 . . . bn lt
x¼4 5
y10 y11c y11s y12 y_ 10 y_ 11c y_ 11s y_ 12 . . . yk0 yk1c yk1s yk2 y_ k0 y_ k1c y_ k1s y_ k2
ð37Þ
where u, v, w is the translational velocity components, p, q, r is
the angular velocity components, f, u, c is the Euler angles,
yk0 ; yk1c ; yk1s ; yk2 are, respectively, the collective, longitudinal,
lateral and differential tilt angle of the k-mode.
The linearization produces matrices of an order equal to the
length of the full state vector.
5. Simulation results
5 In this section, the transient responses of an articulated rotor
helicopter to step-inputs at various trim conditions have been
calculated, and they have compared with flight test data
(Ballin, 1987). The test data used for comparison were obtained
in a series of tests conducted for use in validation of the
Rotorcraft Systems Integration Simulator. The trim data and
selected transient-response time histories were provided to
Sikorsky for use in their validation of the mathematical models.
X  No stability augmentation was used during transient-response
y~ k ðcb Þ ¼ y0k þ yjck cos cb þ yjsk sin c b k ¼ 1; 2; . . . ; 4 ð36Þ
j¼1
The superscript k refers to the kth mode in the modal
coordinate transformation and m ¼ 3 is the highest harmonic
accounted for in the truncated Fourier series.
It should be noted that the trim states vector yielding the
equilibrium position of the helicopter is essentially the base of
the numerical linearization process of the equations of
motion. Here, the equations of motion is written in the
form of the first order differential equations, hence linearized
states and controls matrices are calculated by perturbing the
equations of motion, one state or control at a time produced
by using the finite difference approximations.
Since the linearization process generates the rotor states in
the rotating coordinate system, the multi-blade coordinate
system is applied for transforming those into the non-rotating,
body-fixed coordinate system as:
31
data acquisitions. Analog and digital stability augmentation
systems, the flight path stabilization system, and the horizontal
stabilizer control system were disabled. This is a highly
degraded configuration; the results are not representative of
the UH-60A in normal operation. The test procedure normally
consisted of stabilizing in trim with one of the two redundant
stability augmentation systems on; this was disabled one second
before the control input. Unsatisfactory stability characteristics
of the un-augmented aircraft, especially in pitch, required the
pilot to initiate recovery within a few seconds of the input for
reasons of instability or safety. Furthermore, because the test
program was organized in order to provide standard handling
qualities data, the presented results are focused on both off-axis
and on-axis dynamic response. Validations are discussed in
terms of the pitch, roll and yaw rate responses, which are
important from the handling qualities point of view.
In all cases, trim control settings are normally obtained from
the previous section for hover and 50-knot forward flight speed,
 Indirect responses of helicopters using a refined inflow model Aircraft Engineering and Aerospace Technology: An International Journal
Farid Shahmiri and Fariborz Saghafi Volume 81 · Number 1 · 2009 · 25 –37

corresponding to advanced ratios equaled to 0 and 0.11, Figure 8 Forth natural mode shape
respectively. All results are taken for an altitude of 3,000 ft in a
standard atmosphere and a gross weight of 16,000 lbf. This Frequency = 4.8516/Rev
1
corresponds to a of 0.069. Flap
The current analysis uses six main rotor blade modes to

(Non-Dimensional)
Lag
model the rotor flexibility, including two natural flap modes, 0.5

Displacement
Torsion
two natural lag modes, and two natural torsion modes resulting
from the finite element analysis. Five finite elements used in the 0
calculation of these modes in compromise between
computational cost and accuracy of the simulation. The non- −0.5
uniform blade mass and stiffness distributions are obtained
through lookup tables. The six lowest natural frequencies of −1
modes are corresponded to 0.2650, 1.0351, 2.81, 4.92, 5.797 0 0.2 0.4 0.6 0.8 1
and 8.12 per revelation (Figures 5 through 10). These modes Blade Span (Non-Dimensional)
are calculated with the geometric pitch angle at the root of the
blade set to zero and the centers of gravity and shear of a blade
Figure 9 Fifth natural mode shape
Figure 5 First natural mode shape
Frequency = 5.6797/Rev
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

Frequency = 0.2650/Rev 1
1 Flap

(Non-Dimensional)
0.5 Lag

Displacement
(Non-Dimensional)

Torsion
0.5
Displacement

0
0
Flap −0.5
−0.5 Lag
Torsion −1
−1 0 0.2 0.4 0.6 0.8 1
0 0.2 0.4 0.6 0.8 1 Blade Span (Non-Dimensional)
Blade Span (Non-Dimensional)

Figure 10 Sixth natural mode shape

Figure 6 Second natural mode shape
Frequency = 8.1276/Rev
Frequency = 1.0351/Rev 1.5
1
1
(Non-Dimensional)
(Non-Dimensional)

Displacement

0.5 0.5
Displacement

0
0
−0.5 Flap
Lag Lag
−0.5 Flap −1 Torsion
Torsion
−1.5
−1 0 0.2 0.4 0.6 0.8 1
0 0.2 0.4 0.6 0.8 1
Blade Span (Non-Dimensional)
Blade Span (Non-Dimensional)

coincident at the quarter-chord to reduce the flap-lag and flap-

Figure 7 Third natural mode shape torsion coupling. The presence of the large amount of structural
twist, 2 148, of the blade for this case have a significant influence
Frequency = 2.7817/Rev on the coupling between the flap and lag modes.
1 Flap Lag Torsion The main rotor inflow is modeled by two levels of
qualification (Table I). In the simpler blade model, only the
(Non-Dimensional)

0.5 Pitt-Peters inflow associated with 3 inflow states is used to

Displacement

model the normal induced velocity; in the more refined

0 model, the finite state wake inflow model with 28 wake states
including 16 Cosine and 12 Sine inflow states are used.
−0.5 Furthermore, the blade sectional aerodynamic is also
modeled by two levels of acceptance. From the point of
−1 compatibility with the inflow models, the simpler inflow
0 0.2 0.4 0.6 0.8 1 model is accompanied by unsteady aerodynamic model
Blade Span (Non-Dimensional) presented by Leishman and Nguyen (1990), and the refined

32
 Indirect responses of helicopters using a refined inflow model Aircraft Engineering and Aerospace Technology: An International Journal
Farid Shahmiri and Fariborz Saghafi Volume 81 · Number 1 · 2009 · 25 –37

Table I Summary of the effects captured by sectional aerodynamic and inflow models
Circulatory
Non Shed vortices Trailed Inflow Compressibility Tip loss
Modeling method circulatory (dynamic wake) vortices distribution effects effects
Induced inflow Pitt-Peters (three states) X X U Linear (2D) X X
Finite states (28 states) X U U Non-Linear (3D) X U
Sectional aerodynamic (2D) Quasi-steady U X X – U –
Unsteady U U X – U –

inflow is along with the quasi-steady discussed in this paper. Figure 13 Effect of inflow models on power required versus airspeed
Table I shows two types of modeling considered for
representation of on and off-axis dynamic response of the 3,000
helicopter as a whole. Test Data
The effect of inflow on the various trim results is shown in 2,500 Finite state wake

Power Required (hp)

Figures 11 through 17. Figure 11 shows the result for the Pitt-Peters
collective stick input as a function of forward speed. Consider 2,000
first the simulation results with the three states inflow model.
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

The predicted values of the collective stick agree well with the 1,500
flight test data for speeds greater than about 50 knots, or
m $ 0.1, however, at lower speeds the collective is under 1,000
predicted. The results also show very little effect of blade
500
modeling on the predicted main rotor collective stick, most
likely because of the structural and aerodynamic coupling
0
between the flap, lag and torsional degrees of freedom 0 20 40 60 80 100
(Figure 12). With the finite wake inflow model, the prediction Airspeed (knots)
of the main rotor collective stick (Figure 11) is considerable
improved below 50 knots. This is a results of the higher inflow
predicted by the finite state wake model, compared with that
Figure 14 Effect of blade modeling on helicopter power required
versus airspeed
Figure 11 Effect of inflow models on main rotor collective stick versus
airspeed 3,000
Test Data
Power Required (hp)

20 2,500 Finite state wake-

Main Rotor Collective (Deg)

Test Data 2,000 6th natural modes

Finite state wake Finite state wake-
1,500 2nd natural modes
15 Pitt-Peters
1,000
500
10
0
0 20 40 60 80 100
Airspeed (knots)
5
0 20 40 60 80 100
Airspeed (knots)
Figure 15 Effect of inflow models on main rotor longitudinal cyclic
stick versus airspeed
Figure 12 Effect of blade modeling on main rotor collective stick versus
airspeed 100
Longitudinal Cyclic (Deg)

Test Data
Main Rotor Collective (Deg)

100 80 Finite state wake

Pitt-Peters
80
60
60
40
40
Test Data
20 Finite state wake- 6th natural modes 20
Finite state wake- 2nd natural modes
0 0
0 20 40 60 80 100 0 20 40 60 80 100
Airspeed (knots) Airspeed (knots)

33
 Indirect responses of helicopters using a refined inflow model Aircraft Engineering and Aerospace Technology: An International Journal
Farid Shahmiri and Fariborz Saghafi Volume 81 · Number 1 · 2009 · 25 –37

Figure 16 Effect of inflow models on pitch attitude versus airspeed fail to predict the forward stick required between 20 and 50
knots, which is probably a results of the aerodynamic
10 interactions. For the UH-60A, the incidence of the
Test Data horizontal tail is variable, and is changed by the flight
Finite state wake
Pitch Attitude (Deg)

5 Pitt-Peters
control system as a function of speed. In the present study, the
horizontal tail incidence angle is set to 398 based on the
forward speed which is major cause of deviation in this speed
0 range. The pitch attitude of the fuselage is predicted well
above about 80 knots but is over predicted by a couple of
−5 degrees below this speed.
Figure 17 shows the relation of the lateral stick input for
trim. This figure shows only a small influence of the inflow on
−10 the prediction of the lateral stick position for trim. The
0 20 40 60 80 100
predictions are good for speeds above about 50 knots.
Airspeed (knots)
Figure 18 shows the predicted pedal position. This quantity
is usually coupled to the lateral stick and the required main
rotor power. As it can be seen, the prediction of pedal position
Figure 17 Effect of inflow models on main rotor lateral cyclic stick is improved in the low-speed region with the finite state wake
versus airspeed formulation. This is a result of the higher main rotor torque
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

6
(power required) provided with the finite state wake model,
that is balanced by a higher thrust from the tail rotor. The
Test Data
Finite state wake Bailey solution used is based on momentum theory and a
Lateral Cyclic (Deg)

5 Pitt-Peters uniform inflow distribution that leads to an under prediction

of the collective pitch required to generate the required tail
4 rotor thrust. This may help to explain the under prediction of
the tail rotor collective at all speeds.
Figures 19 and 20 show the correlation of the fuselage roll
3
attitude and sideslip angle, respectively. For V # 50 knots, the
sideslip angle is zero and the trim is performed to a variable
2 roll attitude. A higher fuselage roll attitude to the left is
0 20 40 60 80 100
Airspeed (knots)
Figure 18 Effect of inflow models on tail rotor pedal versus airspeed

predicted by the simple Pitt-Peters inflow model. For speeds 20

from 50 to about 100 knots, the collective stick is slightly over
predicted with the finite state inflow model.
The correlation for the main rotor power required is shown 15
Pedal (Deg)

in Figure 13. The behavior again are similar to those of the

prediction of the collective stick, where below 50 knots, or
m $ 0.1, the power is significantly under predicted using the 10 Test Data
simple dynamic inflow model while the predictions are Finite state wake
significantly improved using the finite state wake model. In Pitt-Peters
this speed range, the majority of the main rotor power
required results from induced effects. Since the finite state 5
0 20 40 60 80 100
wake inflow values are generally higher than those of the
other, a higher power is predicted when using the finite state Airspeed (knots)
wake model. For speeds above 50 knots, the prediction of
the power is improved using the finite state wake model.
As with the prediction of the collective stick, the modeling of Figure 19 Effect of inflow models on roll attitude versus airspeed
the blades does not significantly affect the prediction of the 5
main rotor power required. Test Data
As it can be seen from the Figure 14 the blade modeling, Finite state wake
Roll Attitude (Deg)

changing from second to sixth natural modes shapes, has very Pitt-Peters
little influence on the prediction of power required.
Figures 15 and 16 show the correlation of the longitudinal 0
stick input and fuselage pitch attitude for trim, respectively.
Although the precise prediction of the pitching characteristic
at low speeds is difficult because of the complex aerodynamic
interactions between the main rotor wake and the fuselage
and horizontal tail, both the Pitt-Peters inflow and finite state −5
0 20 40 60 80 100
models show a good prediction of the longitudinal stick
position over the entire speed range. Both models, however, Airspeed (knots)

34
 Indirect responses of helicopters using a refined inflow model Aircraft Engineering and Aerospace Technology: An International Journal
Farid Shahmiri and Fariborz Saghafi Volume 81 · Number 1 · 2009 · 25 –37

Figure 20 Effect of inflow models on sideslip versus airspeed Figure 22 On-axis roll rate response to lateral cyclic input, forward
flight
10
Test Data 20
Finite state wake 15 Test Data
Sideslip Angle (Deg)

5 Pitt-Peters Finite state wake

10
Pitt-Peters
5

Roll rate
0
0
−5
−10
−5 −15
0 20 40 60 80 100
−20
Airspeed (knots) 0 1 2 3 4
Time
predicted with the finite state wake model because of the
higher tail rotor thrust required to balance the main rotor Figure 23 Off-axis yaw rate response to lateral cyclic input, forward
torque in the low-speed region. Both the Pitt-Peters inflow flight
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

and finite state wake model results for the roll attitude agree
well with the flight test data. For speeds above 50 knots, the 20
roll attitude is zero and the trim is performed to a variable 15 Test Data
sideslip angle. The prediction of the fuselage sideslip angle is Finite state wake
improved with the finite state wake model, although in the 10
Pitt-Peters
speed range from 50 to 70 knots, the sideslip angle to the right Yaw rate 5
is over predicted with the finite state wake model. This is a 0
result of the higher main rotor torque associated with the
−5
finite state wake model.
In Figures 21 through 23 the roll, pitch, and yaw rate −10
responses to lateral cyclic input for trimmed flight condition −15
at 50 knots forward speed are shown. The time-history plots
−20
show that the finite state off-axis pitch rate response reaches 0 1 2 3 4
to a maximum value, corresponding to 20.568/s, at which the Time
control and damping moment balances, but with a different
sign and magnitude comparing to flight test data. The
calculated pitch response indicates that the off-axis pitch the off-axis yaw rate response weakly depends on the type of
response still suffers from a problem. This is likely due to the implemented inflow models. This is due to the fact that the
sign of the restoring moment and the helicopter translational inflow only uses the normal induced velocity component.
acceleration, which strongly depend on the induced inflow
distribution over the rotor disc. It seems that there is a phase 6. Conclusions
shift in predicted finite state result in comparison with flight
test data. This is due to the low-frequency nature of the finite The practical implementation of the 3D unsteady induced
state inflow modeling in which it is assumed that the main inflow model, a finite state representation, is included in a
rotor aerodynamics occur slowly. The figures also show that comprehensive flight dynamic simulation program. Based on
the results presented in this paper, the following conclusions
can be drawn; the on-axis responses are not significantly
Figure 21 Off-axis pitch rate response to lateral cyclic input, forward
affected by inflow modeling in the time range of interest for
flight
flight dynamics applications, and correlations with flight test
15 data are generally good. The finite state inflow model, which
Test Data does not include induced wake distortions, shows fairly good
10 Finite state wake predictions of the off-axis responses. The addition of the finite
Pitt-Peters state inflow model fairly improves the off-axis response
5 predictions both quantitatively and qualitatively.
Pitch rate

0
References
−5 Arnold, U.T.P., Keller, J.D. and Curtiss, H.C. (1995), “The
effect of inflow models on the dynamic response of
−10 helicopters”, paper presented at 21st European Rotorcraft
Forum, Saint-Petersburg, Russia, August-September.
−15 Arnold, U.T.P., Keller, J.D., Curtiss, H.C. and Rechert, G.
0 1 2 3 4
(1998), “The effect of inflow models on the predicted
Time response of helicopters”, AHS Journal, Vol. 43, pp. 25-36.

35
 Indirect responses of helicopters using a refined inflow model
Farid Shahmiri and Fariborz Saghafi
Bagai, A., Leishman, J.G. and Park, J.S. (1998),
“Aerodynamic analysis of a helicopter in steady
maneuvering flight using a free-vortex rotor wake model”,
AHS Journal, Vol. 44, pp. 109-20.
Ballin, M.G. (1987), “Validation of a real-time engineering
simulation of the UH-60A helicopter”, NASA TM-88360.
Basset, P.M. (1998), “Study of the rotor wake distortion
effects on the helicopter pitch-roll cross-couplings”,
Proceedings of the 24th European Rotorcraft Forum,
Marseilles, France.
Basset, P.M. and Elomari, A. (1999), “A rotor vortex wake
model for helicopter flight mechanics and its applications to
the prediction of the pitch-up phenomena”, paper
presented at 25th European Rotorcraft Forum, Rome.
Bhagwat, M.J. (2001), “Mathematical modeling of the
transient dynamics of rotor wakes using a time-accurate
free vortex method”, PhD thesis, Department of Aerospace
Engineering, University of Maryland, College Park, MD.
Bhagwat, M.J. and Leishman, J.G. (2001), “Rotor
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)
aerodynamics during maneuvering flight using time-
accurate free-vortex wake”, paper presented at the 57th
Annual Forum of the AHS, Washington, DC.
Celi, R. (2005), “State-space representation of vortex wakes
by the method of lines”, Journal of the American Helicopter
Society, Vol. 50, pp. 195-205.
Chen, R.T.N. (1989), “A survey of non-uniform inflow
models for rotorcraft flight dynamics and control
applications”, Proceedings of the 15th Annual European
Rotorcraft Forum, Amsterdam, The Netherlands.
Fletcher, J.M. and Tischler, M.B. (1997), “Improving
helicopter flight mechanics models with laser
measurements of flapping”, Proceedings: American
Helicopter Society, Virginia Beach, VA, USA.
Gaonkar, G.H. and Peters, D.A. (1986), “Effectiveness of
current dynamic-inflow models in hover and forward
flight”, AHS Journal, Vol. 31 No. 2, pp. 47-57.
Gaonkar, G.H. and Peters, D.A. (1988), “Review of dynamic
inflow modeling for rotorcraft flight dynamics”, Vertica,
Vol. 12 No. 3, pp. 213-42.
He, C. (1989), “Development and application of a
generalized dynamic wake theory for lifting rotors”,
PhD thesis, School of Aerospace Engineering, Georgia
Institute of Technology, Atlanta, GA.
He, C., Lee, C.S. and Chen, W. (1999), “Rotorcraft
simulation model enhancement to support design, testing
and operational analysis”, paper presented at The AHS
55th Annual Forum, Montreal.
Keller, J.D. (1996), “An investigation of helicopter dynamic
coupling using an analytical model”, AHS Journal, Vol. 41
No. 4, pp. 322-30.
Keller, J.D. and Curtiss, H.C. (1996), “Modeling the induced
velocity of a maneuvering helicopter”, paper presented at
the AHS 52nd Annual Forum, Washington, DC.
Keller, J.D. and Curtiss, H.C. (1998), “A critical examination
of the methods to improve the off-axis response prediction
of helicopters”, The American Helicopter 54th Annual Forum,
Washington, DC, pp. 1134-47.
Leishman, J.G. and Nguyen, K.Q. (1990), “State space
representation of unsteady airfoil behavior”, AIAA Journal,
Vol. 28 No. 5, pp. 836-44.
Lovera, M., Colaneri, P., Malpica, C. and Celi, R. (2007),
“Discrete-time, closed-loop aeromechanical stability
36
Aircraft Engineering and Aerospace Technology: An International Journal
Volume 81 · Number 1 · 2009 · 25 –37
analysis of helicopters with higher harmonic control”,
Journal of Guidance, Control, and Dynamics, Vol. 30.
Park, J.S. and Leishman, J.G. (1999), “Investigation of
unsteady aerodynamics on rotor wake effects in
maneuvering flight”, paper presented at The 55th Annual
Forum of the AHS, Montreal.
Peters, D.A. and He, C.J. (1991), “Correlation of measured
induced velocities with a finite-state wake model”,
AHS Journal, Vol. 36 No. 3, pp. 59-70.
Peters, D.A., Boyd, D.D. and He, C.J. (1989), “Finite-state
induced-flow model for rotors in hover and forward flight”,
AHS Journal, Vol. 34 No. 4, pp. 5-17.
Pitt, D.M. and Peters, D.A. (1981), “Theoretical prediction
of dynamic inflow derivatives”, Vertica, Vol. 5 No. 1,
pp. 21-34.
Ribera, M. and Celi, R. (2004), “Simulation modeling of
unsteady maneuvers using a time accurate free wake”,
American Helicopter Society 60th Annual Forum Proceedings,
Baltimore, MD, USA.
Ribera, M. and Celi, R. (2005), “Simulation modeling in
steady turning flight with refined aerodynamics”,
Proceedings of the 31st European Rotorcraft Forum, Firenze,
Italy.
Ribera, M. and Celi, R. (2007a), “Simulation modeling in
climbing and descending flight with refined aerodynamics”,
Proceedings of the 62nd Annual Forum of the American
Helicopter Society, Phoenix, Arizona, USA.
Ribera, M. and Celi, R. (2007b), “Time marching simulation
modeling in axial descending flight through the vortex ring
state”, Proceedings of the 63rd Annual Forum of the American
Helicopter Society, Virginia Beach, VA, USA.
Rosen, A. (2004a), “Approximate actuator disk model of a
rotor in hover or axial flow based on vortex modeling”,
AHS Journal, Vol. 49 No. 1, pp. 66-79.
Rosen, A. (2004b), “Approximate actuator disk model of a
rotor in hover or axial flow based on potential flow
equations”, AHS Journal, Vol. 49 No. 1, pp. 80-92.
Rosen, A. and Isser, A. (1995), “A new model of rotor
dynamics during pitch and roll of a hovering helicopter”,
AHS Journal, Vol. 40 No. 3, pp. 17-28.
Schulein, G. (1999), “Improvements to a high fidelity
helicopter math model”, Master thesis, California
Polytechnic State University, San Luis Obispo, CA.
Shahmiri, F. and Saghafi, F. (2007), “Improvement
of dynamic response prediction of helicopters”,
Aircraft Engineering & Aerospace Technology, Vol. 79,
pp. 579-92.
Shamie, J. and Friedmann, P.P. (1977), “Effect of moderate
deflections on the aeroelastic stability of a rotor blade in
forward flight”, Proceedings of the 3rd Annual European
Rotorcraft and Powered Lift Aircraft Forum, Aix-en-Provence,
France.
Simons, I.A. and Modha, A.N. (2002), “Gyroscopic
feathering moments and the bell stabilizer bar on
helicopter rotors”, paper presented at the 2002 European
Rotorcraft Forum, Bristol.
Theodore, C. and Celi, R. (2003), “Prediction of the off-axis
response to cyclic pitch using a maneuvering free wake
model”, paper presented at the 25th European Rotorcraft
Forum, Rome.
Tischler, M.B. (1999), “Identification of bearing-less main
rotor dynamic characteristics from frequency-response
 Indirect responses of helicopters using a refined inflow model
Farid Shahmiri and Fariborz Saghafi
wind-tunnel test data”, AHS Journal, Vol. 44 No. 1,
pp. 63-76.
von Grünhagen, W. (1995), “Dynamic inflow modeling for
helicopter rotors and its influence on the prediction of
cross-couplings”, Proceedings of the AHS Aeromechanics
Specialists Conference, Bridgeport, CT, USA.
Further reading
Basset, P.M. (1996), “Modeling of the dynamic inflow on the
main rotor and the tail components in helicopter flight
mechanics”, Proceedings of the 22nd European Rotorcraft
Forum, Brighton.
Kinner, W. (1937), “Die kreisförmige tragfläche auf
potential theoretischer grundlage”, Ing. Arch, Vol. 8 No. 1,
pp. 47-80.
Peters, D.A. and HaQuang, N. (1988), “Dynamic inflow for
practical applications”, AHS Journal, Vol. 33 No. 4,
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)
pp. 64-8.
To purchase reprints of this article please e-mail: reprints@emeraldinsight.com
Or visit our web site for further details: www.emeraldinsight.com/reprints
37
Aircraft Engineering and Aerospace Technology: An International Journal
Volume 81 · Number 1 · 2009 · 25 –37
About the authors
Farid Shahmiri obtained his MSc degree in
Aerospace (Aerodynamic field) in 2000 and
currently is a PhD student in Flight Dynamics
Simulation Group in Aerospace Engineering
Department of Sharif University of Technology.
His main research interests are helicopter flight
dynamic simulation, rotor wake modeling,
aeroelasticity, trim, stability and response analyses.
Farid Shahmiri is the corresponding author and can be
contacted at: Farid_shahmiri@yahoo.co.uk
Fariborz Saghafi obtained his PhD in
Flight Dynamics from Cranfield University,
UK, in 1996. He is currently an academic
member of Flight Dynamics Group in
Aerospace Engineering Department of Sharif
University of Technology. His main field of
research is the dynamics, control and simulation
of novel VTOL concepts.
 This article has been cited by:

1. Ranjan Ganguli. 2010. Rotorcraft research in India: recent developments. Aircraft Engineering and Aerospace Technology 82:5,
305-319. [Abstract] [Full Text] [PDF]
Downloaded by UNIVERSITY OF OTAGO At 20:55 25 December 2014 (PT)

View publication stats