Recent Advances in Applications of

Computational and Fuzzy Mathematics

Snehashish Chakraverty
Visit to download the full and correct content document:
https://textbookfull.com/product/recent-advances-in-applications-of-computational-and
-fuzzy-mathematics-snehashish-chakraverty/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Recent Advances in Intuitionistic Fuzzy Logic Systems

and Mathematics Said Melliani

https://textbookfull.com/product/recent-advances-in-
intuitionistic-fuzzy-logic-systems-and-mathematics-said-melliani/

Uncertainty Management with Fuzzy and Rough Sets Recent

Advances and Applications Rafael Bello

https://textbookfull.com/product/uncertainty-management-with-
fuzzy-and-rough-sets-recent-advances-and-applications-rafael-
bello/

Recent Trends in Wave Mechanics and Vibrations Select

Proceedings of WMVC 2018 S. Chakraverty

https://textbookfull.com/product/recent-trends-in-wave-mechanics-
and-vibrations-select-proceedings-of-wmvc-2018-s-chakraverty/

Recent Advances in Computational Optimization: Results

of the Workshop on “Computational Optimization” and
“Numerical Search and Optimization” 2018 Stefka
Fidanova
https://textbookfull.com/product/recent-advances-in-
computational-optimization-results-of-the-workshop-on-
computational-optimization-and-numerical-search-and-
Recent Advances in Computational Engineering
Proceedings of the 4th International Conference on
Computational Engineering ICCE 2017 in Darmstadt
Michael Schäfer
https://textbookfull.com/product/recent-advances-in-
computational-engineering-proceedings-of-the-4th-international-
conference-on-computational-engineering-icce-2017-in-darmstadt-
michael-schafer/

Recent Advances in Intelligent Assistive Technologies

Paradigms and Applications Hariton Costin

https://textbookfull.com/product/recent-advances-in-intelligent-
assistive-technologies-paradigms-and-applications-hariton-costin/

Recent Advances in Computational Mechanics and

Simulations, Volume-II: Nano to Macro Sandip Kumar Saha

https://textbookfull.com/product/recent-advances-in-
computational-mechanics-and-simulations-volume-ii-nano-to-macro-
sandip-kumar-saha/

Recent Advances in Engineering Mathematics and Physics:

Proceedings of the International Conference RAEMP 2019
Mohamed Hesham Farouk

https://textbookfull.com/product/recent-advances-in-engineering-
mathematics-and-physics-proceedings-of-the-international-
conference-raemp-2019-mohamed-hesham-farouk/

Computational Mathematics and Applications 1st Edition

Dia Zeidan

https://textbookfull.com/product/computational-mathematics-and-
applications-1st-edition-dia-zeidan/
Snehashish Chakraverty
Sanjeewa Perera Editors

Recent Advances
in Applications of
Computational
and Fuzzy
Mathematics
Recent Advances in Applications of Computational
and Fuzzy Mathematics
Snehashish Chakraverty Sanjeewa Perera
•

Editors

Recent Advances
in Applications of
Computational and Fuzzy
Mathematics

123
Editors
Snehashish Chakraverty Sanjeewa Perera
Department of Mathematics Research and Development Centre
National Institute of Technology for Mathematical Modeling,
Rourkela Faculty of Science
Rourkela, Odisha, India University of Colombo
Colombo, Sri Lanka

ISBN 978-981-13-1152-9 ISBN 978-981-13-1153-6 (eBook)

https://doi.org/10.1007/978-981-13-1153-6
Library of Congress Control Number: 2018945436

© Springer Nature Singapore Pte Ltd. 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained herein or
for any errors or omissions that may have been made. The publisher remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.

Printed on acid-free paper

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Preface

In general, many real-world problems are inherently biased by uncertainty. There is

often a need to model, solve and interpret the problems one encounters in the world
of uncertainty. Models and parameters of practical application are usually estab-
lished on the basis of measurements, observations, experiences, expert knowledge,
codes and standards, etc. Usually, exact information and precise values do not exist.
The uncertainty in both data and models must be appropriately handled. This
contributed volume includes a collection of investigations on computational and
fuzzy mathematics presenting the recent advances in real-world problems.
The first chapter deals with “2-D Shallow Water Wave Equations with Fuzzy
Parameters” by P. Karunakar and Snehashish Chakraverty which examines the
behavior of two-dimensional shallow water wave equations (SWWEs) with fuzzy
parameters. In this study, basin depth has been considered as uncertain in terms of
fuzzy. Then, the corresponding SWWE has been solved by semianalytical
approach, viz. homotopy perturbation method (HPM).
The second chapter is on “ANN Based Solution of Static Structural Problem
with Fuzzy Parameters” by S. K. Jeswal and Snehashish Chakraverty which
incorporates a new method, viz. the concept of artificial neural network (ANN), in
solving the fuzzy linear system of equations corresponding to the static problem of
structure. The detailed procedure is presented here, followed by simulation for
different example problems of civil structures. The algorithm has also been illus-
trated by solving few numerical examples, and the obtained results are compared in
special cases.
The third chapter addresses the problem entitled “Fuzzy Matrix Contractor
Based Approach for Localization of Robots” by N. R. Mahato, Snehashish
Chakraverty and L. Jaulin. This chapter deals with the localization of a group of
robots using angle and distance constraints associated with fuzzy matrix contrac-
tors. Various groups of robots have been considered for the verification of proposed
contractors, viz. azimuth, distance, azimuth–distance and bearing–distance con-
tractors using Gaussian fuzzy uncertainty.

v
vi Preface

The fourth chapter includes “Modeling Radon Diffusion Equation by Using

Fuzzy Polynomials in Galerkin’s Method” by T. D. Rao and Snehashish
Chakraverty. In this chapter, a new technique based on fuzzy polynomials in
Galerkin’s method has been proposed to solve the uncertain radon diffusion
equation. The shape functions in the initial approximation have been taken as fuzzy
polynomials which satisfy the given boundary conditions. The uncertainty has been
taken in the boundary/initial conditions in fuzzy form. The obtained results are
compared with their exact solutions (wherever possible) in order to demonstrate the
validity and applicability of the method.
The fifth chapter is on “Solving Fuzzy Static Structural Problems Using
Symmetric Group Method” by S. K. Jena and Snehashish Chakraverty. Here, the
static problems that are encountered in civil, mechanical and aerospace structures
with fuzzy parameters are considered which convert the problem into fuzzy dif-
ferential equations. The properties of symmetries provide a unique tool for solving
crisp differential equations. Accordingly, the target of this chapter has been to use
symmetry to solve fuzzy differential equations. The simple example of structural
problems has been solved for showing the powerfulness and reliability of the
method. The obtained solutions are compared with crisp solutions in special cases.
Further, the sixth chapter targets the study entitled “Modeling Dispersal Risk of
Invasive Alien Plant Species” by H. O. W. Peiris, Sanjeewa Perera, Snehashish
Chakraverty and S. M. W. Ranwala. In this chapter, several fuzzy
mathematics-based models have been considered to assess the risk of invasive alien
plant species. The first four models are based on fuzzy set theory operators, whereas
remaining models are developed by using fuzzy analytical hierarchy techniques.
The models are validated using known invasive and noninvasive species.
Finally, the last chapter introduces the investigation entitled “Mathematical
Model to Quantify Air Pollution in Cities” by I. T. S. Piyatilake and Sanjeewa
Perera which presents how to initiate, plan and form a mathematical model to
identify the air pollution level in cities. The concepts of fuzzy analytical hierarchical
process and fuzzy operators are used together to build the model. It also demon-
strates an application of the air quality model to identify the pollution level in urban
areas of Colombo, Sri Lanka.

Rourkela, India Snehashish Chakraverty

Colombo, Sri Lanka Sanjeewa Perera
About the Book

This multi-author contributed volume contains recent advances in several areas of

computational and fuzzy mathematics. This book focuses on handling the uncer-
tainty in terms of fuzzy mathematics and machine intelligence.
This book contains seven chapters covering various aspects of fuzzy uncertainty
from theoretical to application problems. This book is an attempt to bring together
the faculties, scientists, engineers and technologists from various fields of science
and engineering to discuss the recent trends, usefulness and challenges of fuzzy
uncertainty in the following challenging application problems: 2-D Shallow Water
Wave Equation with Fuzzy Parameters; ANN-Based Solution of Static Structural
Problem with Fuzzy Parameters; Fuzzy Matrix Contractor-Based Approach for
Localization of Robots; Modeling Radon Diffusion Equation by Using fuzzy
polynomials in Galerkin’s Method; Solving Fuzzy Static Structural Problems Using
Symmetric Group Method; Modeling Dispersal Risk of Invasive Alien Species;
Mathematical Model to Quantify Air Pollution in Cities.
Recent Advances in Applications of Computational and Fuzzy Mathematics is
aimed for the researchers who are working in these challenging areas.

vii
Contents

1 2-D Shallow Water Wave Equations with Fuzzy Parameters . . . . . . 1

P. Karunakar and Snehashish Chakraverty
2 ANN Based Solution of Static Structural Problem with Fuzzy
Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
S. K. Jeswal and Snehashish Chakraverty
3 Fuzzy Matrix Contractor Based Approach for Localization
of Robots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
N. R. Mahato, Snehashish Chakraverty and L. Jaulin
4 Modeling Radon Diffusion Equation by Using Fuzzy
Polynomials in Galerkin’s Method . . . . . . . . . . . . . . . . . . . . . . . . . . 75
T. D. Rao and Snehashish Chakraverty
5 Solving Fuzzy Static Structural Problems Using Symmetric
Group Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
S. K. Jena and Snehashish Chakraverty
6 Modeling Dispersal Risk of Invasive Alien Plant Species . . . . . . . . . 109
H. O. W. Peiris, Sanjeewa Perera, Snehashish Chakraverty
and S. M. W. Ranwala
7 Mathematical Model to Quantify Air Pollution in Cities . . . . . . . . . . 147
I. T. S. Piyatilake and Sanjeewa Perera

ix
Editors and Contributors

About the Editors

Dr. Snehashish Chakraverty has 27 years of experience as a researcher and

teacher. He is currently a professor of mathematics at the National Institute of
Technology Rourkela, Odisha, India. Prior to this, he was with CSIR—Central
Building Research Institute, Roorkee, India. He received his Ph.D. from IIT
Roorkee in 1992. He then pursued postdoctoral research at the Institute of Sound
and Vibration Research (ISVR), University of Southampton, UK, and at the Faculty
of Engineering and Computer Science, Concordia University, Canada. He was a
visiting professor at Concordia and McGill Universities, Canada, from 1997 to
1999, and visiting professor at the University of Johannesburg, South Africa, in
2011–2014. He has authored/co-authored 11 books and published 283 research
papers in journals and conferences, and two more books are ongoing. He is on the
editorial boards of various international journals, book series and conferences. He is
the chief editor of the International Journal of Fuzzy Computation and Modelling
(IJFCM), Inderscience Publishers, Switzerland, and is guest editor for various other
journals. He is also a reviewer for around 50 national and international journals and
was the president of the Section of Mathematical Sciences (including Statistics) of
Indian Science Congress (2015–2016) and the Vice President of Orissa
Mathematical Society (2011–2013). He is the recipient of a number of prestigious
awards: Indian National Science Academy (INSA) nomination under the
International Collaboration/Bilateral Exchange Program (with Czech Republic),
Platinum Jubilee ISCA Lecture Award (2014), CSIR Young Scientist (1997),
BOYSCAST (DST), UCOST Young Scientist (2007, 2008), Golden Jubilee
Director’s (CBRI) Award (2001), INSA International Bilateral Exchange Award
[2010–2011 (selected but could not undertake), 2015 (selected)] and Roorkee
University Gold Medals (1987, 1988) for first positions in M.Sc. and M.Phil. He
has been the principal investigator of 16 research projects funded by international
and national agencies. His current research areas include soft computing and
machine intelligence, artificial neural networks, fuzzy and interval computations,

xi
xii Editors and Contributors

numerical analysis, differential equations, mathematical modeling, uncertainty

modeling, vibration and inverse vibration problems.
Dr. Sanjeewa Perera has 16 years of research and teaching experience. He
received his Ph.D. in mathematical modeling from the TU Kaiserslautern,
Germany, and University of Colombo, Sri Lanka (2008), under the DAAD
scholarship program, and his M.Sc. from ICTP/SISSA, Trieste, Italy (2004). He is
currently working at the University of Colombo as a senior lecturer in mathematics.
His research interests include mathematical modeling in biology, infectious dis-
eases, natural sciences, actuarial science, computational biology, fuzzy logic-based
modeling and stochastic modeling. He has published numerous chapters as well as
research papers in various leading national and international journals. He is a
member of the Editorial Board of the Journal of National Science Foundation,
which is the only Sri Lankan journal in the science citation index.

Contributors

Snehashish Chakraverty Department of Mathematics, National Institute of

Technology Rourkela, Rourkela, Odisha, India
L. Jaulin CNRS 6285, LabSTICC, ENSTA-Bretagne, Brest, France
S. K. Jena Department of Mathematics, National Institute of Technology
Rourkela, Rourkela, Odisha, India
S. K. Jeswal Department of Mathematics, National Institute of Technology
Rourkela, Odisha, India
P. Karunakar Department of Mathematics, National Institute of Technology,
Rourkela, Odisha, India
N. R. Mahato Department of Mathematics, National Institute of Technology
Rourkela, Rourkela, Odisha, India
H. O. W. Peiris Department of Mathematics, Faculty of Natural Science, The
Open University of Sri Lanka, Nawala, Nugegoda, Sri Lanka
Sanjeewa Perera Research & Development Centre for Mathematical Modeling,
Faculty of Science, University of Colombo, Colombo, Sri Lanka
I. T. S. Piyatilake Department of Computational Mathematics, Faculty of
Information Technology, University of Moratuwa, Moratuwa, Sri Lanka
S. M. W. Ranwala Department of Plant Sciences, Faculty of Science, University
of Colombo, Colombo, Sri Lanka
T. D. Rao Department of Mathematics, National Institute of Technology
Rourkela, Rourkela, Odisha, India
Chapter 1
2-D Shallow Water Wave Equations
with Fuzzy Parameters

P. Karunakar and Snehashish Chakraverty

1.1 Introduction

Water waves, with much longer horizontal scale of flow than the depth of the fluid,
are considered as shallow water waves. SWWEs are widely used for simulating the
tsunami waves because tsunamis are examples of such waves whose horizontal flow
is larger than depth. As demonstrated by Carrier and Greenspan in [1] and Hibberd
and Peregrine in [2], a correct estimation of levels and velocity fields in the very
shallow and transitional dry regions is a difficult numerical problem because strong
velocity gradients occur near the run-up point. Due to nonlinearity, hyperbolicity of
the shallow water equations, and other difficulties, these equations have applications
to a wide range of phenomena other than water waves, e.g., avalanches and atmo-
sphere flow. In the case of free surface flow when the shallow water approximation
is not valid, it is common to model the surface waves using several layers of shallow
water equations coupled via the pressure [3–5]. Many researchers proposed various
methods to solve SWWE in crisp environment. Cho et al. [6] proposed a numerical
method for describing a distance propagation of linear shallow water waves equa-
tions. The applications of linear and nonlinear SWWE in tsunami simulations are
studied by Liu et al. [7]. Numerical methods of tsunami simulations using leap-frog
scheme are presented in [8]. Exp-function method has been applied by Bekir and
Esin [9] for finding the solution of extended SWWE. Safari and Safari [10] used
He’s variational iteration method (VIM) to find the analytical solution of shallow
water waves governed by two extended model equations. Various numerical meth-
ods for shallow water flow may be found in Vreugdenhil [11]. It may be noted that

P. Karunakar · S. Chakraverty (B)

Department of Mathematics, National Institute of Technology,
Rourkela 769008, Odisha, India
e-mail: sne_chak@yahoo.com; chakravertys@nitrkl.ac.in
P. Karunakar
e-mail: karunakarperumandla@gmail.com
© Springer Nature Singapore Pte Ltd. 2018 1
S. Chakraverty and S. Perera (eds.), Recent Advances in Applications
of Computational and Fuzzy Mathematics,
https://doi.org/10.1007/978-981-13-1153-6_1
2 P. Karunakar and S. Chakraverty

we may not always have exact information about the parameters involved in the gov-
erning equations of shallow water wave and those may be uncertain. As such here,
fuzzy parameters are introduced to handle the governing equations of SWWE. To
the best of our knowledge, no work has been reported to solve SWWE with fuzzy
parameters. Accordingly, present work aims to solve SWWE with fuzzy parameters
using the efficient method, viz. homotopy perturbation method [12]. HPM can be
used for solving linear and nonlinear differential equations and coupled differential
equations. As such, He [13, 14] introduced coupling method of HPM to solve linear
and nonlinear partial and ordinary differential equations. Sheikholeslami et al. [15]
used HPM for investigating micropolar fluid flow in a channel subject to a chemical
reaction. Recently, coupled 1D SWWE with Gaussian-type initial conditions have
been solved using HPM by Karunakar and Chakraverty [16]. A new modification
to homotopy perturbation method has been introduced by Singh et al. [17] for solv-
ing nonlinear and singular time-dependent Emden–Fowler-type equations with the
Neumann and Dirichlet boundary conditions. The convergence of HPM has been
discussed by Ayati and Biazar [18] using Banach’s fixed point theorem. Tapaswini
and Chakraverty [19] proposed a new technique based on double parametric from
of fuzzy numbers for solving uncertain beam equation with the help of Adomian
decomposition method. Further, Chakraverty et al. [20] extended the use of varia-
tional iteration method and HPM for solving fuzzy ordinary, partial, and fractional
order differential equations.

1.2 Preliminaries of Shallow Water Wave Equations

Here we give brief introduction about the coupled linear and nonlinear SWWE [6,
7] in the crisp form. A surface wave whose wavelength is much larger than the depth
of the water is said to be in shallow water. The flow at the free surface of a body due
to the force of gravity or below the horizontal pressure surface is referred as shallow
water wave. SWWEs are a set of partial differential equations that describe shallow
water waves. In order to analyze the tidal oscillations or to understand wave damages
due to tsunamis/storm waves, the governing equations may be that of shallow water
equation. On the other hand, broken wave propagation, sediment transportation,
beach evolution, etc., may also be modeled through SWWE.
The coupled linear SWWE in two dimensions [6] may be given as
∂η ∂ M ∂ N
+ + 0
∂t ∂x ∂y
∂M ∂η
+ gh 0
∂t ∂x
∂N ∂N
+ gh 0 (1.1)
∂t ∂y
1 2-D Shallow Water Wave Equations with Fuzzy Parameters 3

where η represents the water surface elevation (amplitude), t represents the time, M
and N represent the depth—averaged fluxes in x and y directions, g represents the
acceleration due to gravity, and h represents the basin depth.
As such, the coupled nonlinear SWWE in 2D [6, 7] may be written as
∂η ∂ M ∂ N
+ + 0
∂t ∂x ∂y
∂M ∂η
+ g(h + η) 0
∂t ∂x
∂N ∂η
+ g(h + η) 0 (1.2)
∂t ∂y

1.3 Preliminaries of Fuzzy Sets

Here, we present basic fuzzy concepts and notations which are used in further dis-
cussion.
Fuzzy Set: A fuzzy set à is set of pairs consisting of the elements x of a universal
set X and a certain degree of pre-assumed membership values μ(x) of the form

à  {(x, μ(x)) : x ∈ X, μ(x) ∈ [0, 1]}

where μ(x) is the membership function of the fuzzy set and it is piecewise continuous.
Fuzzy Number: A fuzzy set à defined over real line is called a fuzzy number if
à is
(i) convex
(ii) normalized
(iii) piecewise continuous
Types of Fuzzy Numbers
Based on the definition of membership function, there are major four types of
fuzzy numbers as defined below.
(i) Triangular Fuzzy Number
A fuzzy number à [a, b, c] is said to be triangular fuzzy number (TFN) if its
membership function is given as [21],
⎧
⎪
⎪ 0, x ≤b−a
⎪
⎨ 1 + (x − b)/a, b − a ≤ x ≤ b
μ(x) 
⎪
⎪ 1 − (x − b)/c, b ≤ x ≤ b + c
⎪
⎩
0, x ≥b+c
4 P. Karunakar and S. Chakraverty

(ii) Exponential Fuzzy Number

A fuzzy number à [a, b, c] whose membership function is defined as below

[21] is called exponential fuzzy number (EFN).
⎧  
⎨ exp − (x−b) , for x < b
a
μ Ã (x)  ∀x ∈ R
⎩ exp− (x−b) , for x ≥ b
c

(iii) Quadratic Fuzzy Number

A fuzzy number à [a, b, c] is said to quadratic fuzzy number (QFN) if it has

membership function as [21]
⎧
⎪
⎪ 0, x ≤b−a
⎪
⎪
⎨ 1 − (x − b)2 /a 2 , b − a ≤ x ≤ b
μ Ã (x) 
⎪ 1 − (x − b)2 /c2 , b ≤ x ≤ b + c
⎪
⎪
⎪
⎩ 0, x ≥b+c

(iv) Gaussian Fuzzy Number

The membership function of the Gaussian fuzzy number (GFN) is [21]

⎧ 
⎪
⎪ 2
⎨ exp − (x−b)
2a 2
, for x < b
μ Ã (x)   ∀x ∈ R
⎪
⎪ 2
⎩ exp − (x−b)
2c 2 , for x ≥ b

The plots of the above four types of fuzzy numbers are shown in Fig. 1.1.
One may use above-defined four fuzzy numbers TFN, EFN, QFN, and GFN to
handle uncertain parameters present in the governing equations. In this chapter, only
the triangular fuzzy number (TFN) has been used to handle the uncertain basin depth
(h).
r-cut: Using r-cut, any fuzzy number of the form à [a, b, c] may be represented
in the interval form as below [19, 20]

à  [a, b, c]  [b − a + ra, b + c − r c], r ∈ [0, 1]

Parametric Concept: Parametric concept is useful to convert an interval to crisp

form. Accordingly, the interval x̃  [x, x] may be written [19, 20] as

x̃  x + 2βx̃

where β ∈ [0, 1] is a parameter and

1 2-D Shallow Water Wave Equations with Fuzzy Parameters 5

Fig. 1.1 Types of fuzzy numbers

(x − x)
x̃ 
2

1.4 Homotopy Perturbation Method

Now, we present brief idea of HPM proposed by He [12–14].

Let us consider the general nonlinear differential equations

A(u) − f (r )  0, r ∈  (1.3)

subject to boundary conditions

6 P. Karunakar and S. Chakraverty

∂u
B u,  0, r ∈  (1.4)
∂r

where A be the general differential operator, B be the boundary operator,  be the

boundary of the domain , and the function f (r ) is an analytical.
Next, the operator A can be decomposed into two parts; one is linear (L) and
other one is nonlinear (N ). So, Eq. (1.3) may be written as

L(u) + N (u) − f (r )  0. (1.5)

Using homotopy technique, one may construct a homotopy v(r, p) : ×[0, 1] →

R which satisfies

H (v, p)  (1 − p)[L(v) − L(u 0 )] + p[A(v) − f (r )]  0, (1.6)

where p ∈ [0, 1] is an embedding parameter and u 0 is the given condition.

We consider the solution of Eq. (1.6) as a power series in p as below

v  v0 + pv1 + p 2 v2 + · · · (1.7)

The approximate solution of Eq. (1.3) will be obtained by putting p  1 in

Eq. (1.7)

u  lim v  v0 + v1 + v2 + · · · (1.8)
p→1

1.4.1 Convergence of HPM Solution

The convergence analysis has been discussed in this head, with the help of a theorem
proposed by Biazar and Ayati [11].

Theorem 1.1 Let P and Q are two Banach’s spaces and N be a contraction mapping
from P to Q

∀v, ṽ ∈ P; N (v) − N (ṽ) ≤ γ v − ṽ, 0 < γ < 1,

which according to Banach’s fixed point theorem, having the fixed point u, that is
N (u)  u.
The sequence generated by the HPM is considered as
n−1
Vn  N (Vn−1 ), Vn−1  u i , n  1, 2, 3, . . . ,
i0
1 2-D Shallow Water Wave Equations with Fuzzy Parameters 7

and let V0  v0  u 0 ∈ Br (u) where Br (u)  {u ∗ ∈ X |u ∗ − u < r }, then we

have the following statements:
(i) Vn − u ≤ γ n v0 − u,
(ii) Vn ∈ Br (u),
(iii) limn→∞ Vn  u.

Proof may be found in [11], but we are reproducing the same for ready reference.
(i) We may easily prove part (i) using the concept of induction on n,
For n  1 we have

V1 − u  N (V0 ) − N (u)  N (v0 ) − N (u) ≤ γ v0 − u

Assume an induction hypothesis as Vn−1 − u ≤ γ n−1 v0 − u, then

Vn − u  N (Vn−1 ) − N (u)

≤ γ Vn−1 − u ≤ γ γ n−1 v0 − u  γ n v0 − u

(ii) Now using (i),

Vn − u ≤ γ n v0 − u ≤ γ n r < r ⇒ Vn ∈ Br (u)

(iii) Since 0 < γ < 1, limn→∞ γ n  0 and Vn − u ≤ γ n v0 − u,

⇒ limn→∞ Vn − u  0
⇒ limn→∞ Vn  u

1.5 Solution of 2D Coupled Shallow Water Wave Equations

In this section,
√ we have applied HPM to solve 2D SWWE with constant (crisp) basin
depth h  2/10.
Applying HPM to Eq. (1.1), we may get the below

∂η ∂η0 ∂η0 ∂ M ∂ N
 −p + +
∂t ∂t ∂t ∂x ∂y

∂M ∂ M0 ∂ M0 ∂η
 −p + gh
∂t ∂t ∂t ∂x

∂N ∂ N0 ∂ N0 ∂η
 −p + gh (1.9)
∂t ∂t ∂t ∂y

Next we assume solution of the SWWE (1.9) in the form of power series as
8 P. Karunakar and S. Chakraverty

∞
η p k ηk
k0
∞
M p k Mk
k0
∞
N p k Nk (1.10)
k0

By substituting the power series solution (1.10) in Eq. (1.9) and comparing p 0
coefficients, we may obtain
∂η0 ∂η0

∂t ∂t
∂ M0 ∂ M0

∂t ∂t
∂ N0 ∂ N0
 (1.11)
∂t ∂t
Integrating Eq. (1.11) with respect to t, we get
(x−20)2 2
− (y−20)
η0  0.5 e− 10 20

(x−20)2 2
− (y−20)
M0  50 e− 10 20

(x−20)2 2
− (y−20)
N0  50 e− 10 20

which are the given initial conditions.

Comparing p 1 coefficients, we have

∂η1 ∂η0 ∂ M0 ∂ N0
− + +
∂t ∂t ∂x ∂y

∂ M1 ∂ M0 ∂η0
− + gh
∂t ∂t ∂x

∂ N1 ∂ N0 ∂η0
− + gh (1.12)
∂t ∂t ∂y

Comparing p 2 coefficients, one may obtain


∂η2 ∂ M1 ∂ N 1
− +
∂t ∂x ∂y
∂ M2 ∂η1
 −gh
∂t ∂x
∂ N2 ∂η1
 −gh (1.13)
∂t ∂y
1 2-D Shallow Water Wave Equations with Fuzzy Parameters 9

Comparing p 3 coefficients, we get


∂η3 ∂ M2 ∂ N 2
− +
∂t ∂x ∂y
∂ M3 ∂η2
 −gh
∂t ∂x
∂ N3 ∂η2
 −gh (1.14)
∂t ∂y

In general, one can write as


∂ηn+1 ∂ Mn ∂ N n
− +
∂t ∂x ∂y
∂ Mn+1 ∂ηn
 −gh
∂t ∂x
∂ Nn+1 ∂ηn
 −gh
∂t ∂y

Integrating Eqs. (1.12)–(1.14), we get

−1 −1 (x−20)2
− (y−20)
2
η1  − 50 x + 4 + 50 y+2 e− 10 20 t
5 10

49 √ −1 (x−20)2 (y−20)2
M1  − 2 x + 4 e− 10 − 20 t
100 5

49 √ −1 (x−20)2 (y−20)2
N1  − 2 y + 2 e− 10 − 20 t;
100 10

√ 147 49 −1 49 −1 (x−20)2
− (y−20)
2
η2  2 − + x +4 + y+2 × e− 10 20 t2
2000 200 5 200 10
 
49 √ −1 2
−1 −1
M2  2 −10 + 50 x +4 + 50 y+2 x +4
100 5 10 5
(x−20)2 2
− (y−20)
× e− 10 20 t2

 
49 √ −1 2
−1 −1
N2  2 −5e + 50 y+2 + 50 y+2 x +4
100 10 10 5
(x−20)2 2
− (y−20)
× e− 10 20 t 2;
10 P. Karunakar and S. Chakraverty

 
49 √ −1 −1 3 −1 −1 −1 2
η3  − 2 −30 x + 4 + 50 x + 4 − 10 y + 2 +50 y+2 x +4
300 5 5 10 10 5
  
49 √ −1 −1 −1 2 49 √ −1
− 2 −5 x + 4 + 50 y+2 x +4 − 2 −5 x +4
300 5 10 5 300 5

−1 2 −1 −1 −1 3 (x−20)2 (y−20)2
+50 y+2 x + 4 − 15 y + 2 + 50 y+2 e− 10 − 20 t 3
10 5 10 10


49 √ −343 √ −1 49 √ −1 3
49 √ −1 2
M3  − 2 2 x +4 + 2 x +4 + 2 y+2
50 6000 5 600 5 600 10
−1 (x−20)2 2
− (y−20)
× x +4 e− 10 20 t3
5

49 √ −49 √ −1 49 √ −1 2
−1 49 √
N3  − 2 2 y+2 + 2 x +4 y+2 + 2
50 1200 10 600 5 10 600

3
−1 (x−20)2 (y−20)2
y+2 e− 10 − 20 t 3
10
..
.

Finally, the water surface elevation η of SWWE (1) may be obtained in the series
as η  η0 + η1 + η2 + η3 + · · ·

−1 −1 (x−20)2 (y−20)2 √ 147 49 −1

η  − 50 x + 4 + 50 y+2 e− 10 − 20 t+ 2 − + x +4
5 10 2000 200 5

49 −1 (x−20)2 (y−20)2 49 √ −1 −1 3
+ y+2 e− 10 − 20
+ −t2 2 −30 x + 4 + 50 x +4
200 10 300 5 5
 
−1 −1 −1 2 49 √ −1
− 10 y + 2 +50 y+2 x +4 − 2 −5 x +4
10 10 5 300 5

−1 2 −1 −1 −1 3 (x−20)2 (y−20)2
+50 y+2 x + 4 − 15 y + 2 + 50 y+2 e− 10 − 20 t 3 + · · ·
10 5 10 10

1.6 Shallow Water Wave Equations with Fuzzy Basin Depth

The factors like basin depth, initial wave motion, boundary conditions, and environ-
mental conditions affect the wave motion. Small change in these factors will change
wave motion, which leads to uncertain results. This uncertainty may be modeled
well by considering parameters involved in governing equations as interval or fuzzy
numbers. As such, the basin depth h has been considered here as uncertain in terms
1 2-D Shallow Water Wave Equations with Fuzzy Parameters 11

Fig. 1.2 Triangular fuzzy number h̃  [0.1, 0.1, 0.1]

of fuzzy number, and we assume it as a TFN, viz. h̃  [0.1, 0.1, 0.1]. The TFN plot
of h̃ is shown in Fig. 1.2.
The fuzzy form of the SWWE may be given as

∂ η̃ ∂ M̃ ∂ Ñ
+ + 0
∂t ∂x ∂y
∂ M̃ ∂ η̃
+ g h̃ 0
∂t ∂x
∂ Ñ ∂ Ñ
+ g h̃ 0 (1.15)
∂t ∂y

subject to the initial conditions

(x−20)2 2
− (y−20)
η̃(x, y, 0)  0.5 e− 10 20

(x−20)2 2
− (y−20)
M̃(x, y, 0)  50 e− 10 20

(x−20)2 2
− (y−20)
Ñ (x, y, 0)  50 e− 10 20 (1.16)
12 P. Karunakar and S. Chakraverty

1.7 Solution of Shallow Water Wave Equations with Fuzzy

Basin Depth

In this section, we determine the solution method to solve of the coupled two-
dimensional SWWE (1.15) with a TFN basin depth using homotopy perturbation
method. Before applying HPM, r –cut is used to convert SWWE with fuzzy param-
eters to interval form and then parametric concept [18, 19] has been used to convert
interval form to crisp form. Finally, crisp form of SWWE with fuzzy parameters is
solved using HPM.
The SWWE with fuzzy parameters can also be written as

η̃t (x, y, t) + M̃x (x, y, t) + Ñ y (x, y, t)  0

M̃t (x, y, t) + g h̃ η̃x (x, y, t)  0
Ñt (x, y, t) + g h̃ η̃ y (x, y, t)  0 (1.17)

Here suffixes indicate partial differentiation with respect to the respective variables
and˜ denote the fuzzy form.
First we use r −cut to convert fuzzy differential Eq. (1.17) to interval form as

[ηt (x, y, t; r ), ηt (x, y, t; r )] + [M x (x, y, t; r ), M x (x, y, t; r )]

+ [N y (x, y, t; r ), N y (x, y, t; r )]  0
[M t (x, y, t; r ), M t (x, y, t; r )] + g[h(r ), h(r )][η x (x, y, t; r ), η x (x, y, t; r )]  0
[N t (x, y, t; r ), N t (x, y, t; r )] + g[h(r ), h(r )][η y (x, y, t; r ), η y (x, y, t; r )]  0
(1.18)

Next, interval differential Eq. (1.18) may be written in crisp form using parametric
concept as below

β(ηt (x, y, t; r ) − ηt (x, y, t; r )) + ηt (x, y, t; r ) + β(M x (x, y, t; r ) − M x (x, y, t; r ))

+ M x (x, y, t; r ) + β(N y (x, y, t; r ) − N y (x, y, t; r )) + N y (x, y, t; r )  0
β(M t (x, y, t; r ) − M t (x, y, t; r )) + M t (x, y, t; r )
+ g[β(h(r ) − h(r )) + h(t; r )][β(η x (x, y, t; r ) − η x (x, y, t; r )) + η x (x, y, t; r )]  0
β(N t (x, y, t; r ) − N t (x, y, t; r )) + Nt (x, y, t; r )
+ g[β(h(r ) − h(r )) + h(t; r )][β(η y (x, y, t; r ) − η y (x, y, t; r )) + η y (x, y, t; r )]  0
(1.19)

subject to initial condition:

1 2-D Shallow Water Wave Equations with Fuzzy Parameters 13

(x−20)2 2
− (y−20)
η̃(x, y, 0; r, β)  0.5 e− 10 20

(x−20)2 2
− (y−20)
M̃(x, y, 0; r, β)  50 e− 10 20

(x−20)2 2
− (y−20)
Ñ (x, y, 0; r, β)  50 e− 10 20 (1.20)

where β ∈ [0, 1] is a parameter defined in Sect. 1.3.

Let us denote

β(ηt (x, y, t; r ) − ηt (x, y, t; r )) + ηt (x, y, t; r )  η̃t (x, y, t; r, β)

β(M t (x, y, t; r ) − M t (x, y, t; r )) + M t (x, y, t; r )  M̃t (x, y, t; r, β)
β(N t (x, y, t; r ) − N t (x, y, t; r )) + N t (x, y, t; r )  Ñt (x, y, t; r, β)
β(h(r ) − h(r )) + h(t; r )  h̃(r, β) (1.21)

Using notation defined in (1.21), we rewrite Eq. (1.19) as below

η̃t (x, y, t; r, β) + M̃x (x, y, t; r, β) + Ñ y (x, y, t; r, β)  0

M̃t (x, y, t; r, β) + g h̃(r, β)η̃x (x, y, t; r, β)  0
Ñt (x, y, t; r, β) + g h̃(r, β)η̃ y (x, y, t; r, β)  0 (1.22)

It is worth mentioning here that r and β control the fuzzy solution.

Now, we apply HPM to Eq. (1.22) which involves the parameters r and β. We
construct a homotopy for Eq. (1.22) in the following form

η̃t (x, y, t; r, β)  η̃t (x, y, 0; r, β) − p{η̃t (x, y, 0; r, β) + M̃x (x, y, t; r, β)

+ Ñ y (x, y, t; r, β)}
M̃t (x, y, t; r, β)  M̃t (x, y, 0; r, β) − p{ M̃t (x, y, 0; r, β) + g h̃(r, β)η̃x (x, y, t; r, β)}
Ñt (x, y, t; r, β)  Ñt (x, y, t; r, β) − p{ Ñt (x, y, 0; r, β) + g h̃(r, β)η̃ y (x, y, t; r, β)}
(1.23)

Assume the power series solutions as

η̃(x, y, t; r, β)  η̃0 (x, y, t; r, β) + p η̃1 (x, y, t; r, β)

+ p 2 η̃2 (x, y, t; r, β) + · · ·
M̃(x, y, t; r, β)  M̃0 (x, y, t; r, β) + p M̃1 (x, y, t; r, β)
+ p 2 M̃2 (x, y, t; r, β) + · · ·
Ñ (x, y, t; r, β)  Ñ0 (x, y, t; r, β) + p Ñ1 (x, y, t; r, β)
+ p 2 Ñ2 (x, y, t; r, β) + · · · (1.24)

Next, substitute the power series solutions (1.24) in Eq. (1.23) and compare various
powers of p.
14 P. Karunakar and S. Chakraverty

Comparing p 0 coefficients, we may obtain

∂ ∂
η̃0 (x, y, t; r, β)  η̃(x, y, 0; r, β)
∂t ∂t
∂ ∂
M̃0 (x, y, t; r, β)  M̃(x, y, 0; r, β)
∂t ∂t
∂ ∂
Ñ0 (x, y, t; r, β)  Ñ0 (x, y, 0; r, β) (1.25)
∂t ∂t
Integrating above equations in (1.25) with respect to t, we may obtain
(x−20)2 2
− (y−20)
η̃0 (x, y, t; r, β)  0.5 e− 10 20

(x−20)2 2
− (y−20)
M̃0 (x, y, t; r, β)  50 e− 10 20

(x−20)2 2
− (y−20)
Ñ0 (x, y, t; r, β)  50 e− 10 20 (1.26)

Similarly comparison of p 1 coefficients leads to

∂ ∂ ∂
η̃1 (x, y, t; r, β)  − η̃0 (x, y, t; r, β) + M̃0 (x, y, t; r, β)
∂t ∂t ∂x

∂
+ Ñ0 (x, y, t; r, β)
∂y

∂ ∂ ∂
M̃1 (x, y, t; r, β)  − M̃0 (x, y, t; r, β) + g h̃(r, β) η̃0 (x, y, t; r, β)
∂t ∂t ∂x

∂1 ∂ ∂
Ñ1 (x, y, t; r, β)  − Ñ0 (x, y, t; r, β) + g h̃(r, β) η̃0 (x, y, t; r, β) (1.27)
∂t ∂t ∂y

Comparing p 2 coefficients, we have


∂ ∂ ∂
η̃2 (x, y, t; r, β)  − M̃1 (x, y, t; r, β) + Ñ1 (x, y, t; r, β)
∂t ∂x ∂y
∂ ∂
M̃2 (x, y, t; r, β)  −g h̃(r, β) η̃1 (x, y, t; r, β)
∂t ∂x
∂ ∂
Ñ2 (x, y, t; r, β)  −g h̃(r, β) η̃1 (x, y, t; r, β) (1.28)
∂t ∂y

Similarly, comparing p 3 coefficients one may obtain

1 2-D Shallow Water Wave Equations with Fuzzy Parameters 15

∂ ∂ ∂
η̃3 (x, y, t; r, β)  − M̃2 (x, y, t; r, β) + Ñ2 (x, y, t; r, β)
∂t ∂x ∂y
∂ ∂
M̃3 (x, y, t; r, β)  −g h̃(r, β) η̃2 (x, y, t; r, β)
∂t ∂x
∂ ∂
Ñ3 (x, y, t; r, β)  −g h̃(r, β) η̃2 (x, y, t; r, β) (1.29)
∂t ∂y

and so on.
In general, we may write

∂ ∂ ∂
η̃n+1 (x, y, t; r, β)  − M̃n (x, y, t; r, β) + Ñn (x, y, t; r, β)
∂t ∂x ∂y
∂ ∂
M̃n+1 (x, y, t; r, β)  −g h̃(r, β) η̃n (x, y, t; r, β)
∂t ∂x
∂1 ∂
Ñn+1 (x, y, t; r, β)  −g h̃(r, β) η̃n (x, y, t; r, β), n  1, 2, 3, . . . (1.30)
∂t ∂y

The solution may be obtained by integrating Eqs. (1.27)–(1.30) with respect to t,

and finally, we have the solution of (1.22) as

η̃(x, y, t; r, β)  η̃0 (x, y, t; r, β) + η̃1 (x, y, t; r, β) + η̃2 (x, y, t; r, β)

+ η̃3 (x, y, t; r, β), . . . (1.31)

Next we discuss the convergence of the HPM solution of SWWEs with fuzzy
parameters with help of a theorem.

1.7.1 Convergence of HPM Solution for Shallow Water

Equations with Fuzzy Initial Condition

The partial sums of the series solution (1.31) may be written as [18]

s̃0  η̃0 (x, y, t; r, β)

s̃1  η̃0 (x, y, t; r, β) + η̃1 (x, y, t; r, β)
..
.
s̃n  η̃0 (x, y, t; r, β) + η̃1 (x, y, t; r, β) + η̃2 (x, y, t; r, β) + · · · + η̃n (x, y, t; r, β)
∞
s̃(x, y, t; r, β)  i0 η̃i (x, y, t ; r, β) satisfies the expression

s̃(x, y, t; r, β)  −L −1 (N (s̃(x, y, t; r, β) + η̃0 (x, y, t; r, β))

− η̃(x, y, 0; r, β)) + L( f (r )) (1.32)
16 P. Karunakar and S. Chakraverty

where Eq. (1.32) is equivalent to Eq. (1.6) in terms of fuzzy parameters.

∞
Theorem 1.2 Let B be a Banach’s space. The series solution i0 η̃i (x, y, t; r, β)
convergence to s(x, y, t; r, β) ∈ B, if there exists λ ∈ [0, 1) such that ∀n ∈ N ⇒
η̃n (x, y, t; r, β) ≤ η̃n−1 (x, y, t; r, β). Statement and proof of the above Theorem
for one-dimensional case have been given in [22]. The same is extended here for the
two-dimensional case for the sake of completeness.

Proof Consider,

s̃n+1 (x, y, t; r, β) − s̃n (x, y, t; r, β)

 {η̃0 (x, y, t; r, β) + η̃1 (x, y, t; r, β) + η̃2 (x, y, t; r, β) + · · · + η̃n (x, y, t; r, β)
+ η̃n+1 (x, y, t; r, β)} − {η̃0 (x, y, t; r, β) + η̃1 (x, y, t; r, β) + η̃2 (x, y, t; r, β)
+ · · · + η̃n (x, y, t; r, β)
 η̃n+1 (x, y, t; r, β)
≤ λη̃n (x, y, t; r, β)
≤ λ2 η̃n−1 (x, y, t; r, β)
≤ λn+1 η̃0 (x, y, t; r, β)

For any p, q ∈ N , p ≥ q
 
s̃ p (x, y, t; r, β) − s̃q (x, y, t; r, β)

q  (s̃ p (x, y, t; r, β) − s̃ p−1 (x, y, t; r, β)) + (s̃ p−1 (x, y, t; r, β)

−s̃ p−2 (x, y, t; r, β)) + · · · + (s̃q+1 (x, y, t; r, β) − s̃q (x, y, t; r, β))
  
≤ s̃ p (x, y, t; r, β) − s̃ p−1 (x, y, t; r, β) + s̃ p−1 (x, y, t; r, β)
  
−s̃ p−2 (x, y, t; r, β) + · · · + s̃q+1 (x, y, t; r, β) − s̃q (x, y, t; r, β)
≤ λ p η̃0 (x, y, t; r, β) + λ p−1 η̃0 (x, y, t; r, β) + · · ·
+ λq+1 η̃0 (x, y, t; r, β)
≤ (λ p + λ p−1 + · · · + λq+1 )η̃0 (x, y, t; r, β)
≤ (λq+1 + λq+2 + · · · + λ p + · · ·)η̃0 (x, y, t; r, β)
≤ λq+1 (1 + λ + λ2 · · ·)η̃0 (x, y, t; r, β)
λq+1
≤ η̃0 (x, y, t; r, β)
1−λ
 
As p, q → ∞, λq+1 → 0, then lim s̃ p (x, y, t; r, β) − s̃q (x, y, t; r, β)  0.
p,q→∞
Therefore, the sequence {s̃n (x, y, t; r, β)} is a Cauchy sequence in Banach’s space
and it is convergent.
1 2-D Shallow Water Wave Equations with Fuzzy Parameters 17

1.8 Numerical Results and Discussion

In this section, results obtained for two-dimensional coupled shallow water equations
in both cases, viz. crisp and fuzzy, have been presented.
√Figure 1.3 represents term-wise solutions obtained by HPM at fixed basin depth h
 2/10 and time t  2 s. The results of water surface elevation η of 2D SWWE with
crisp basin depth for time t  2.5 s and 3 s are depicted in Fig. 1.4. Further, Fig. 1.5
represents term-wise plots of the solution (center solution for r  1 and β  0) with
fuzzy basin depth h̃  [0.1, 0.1, 0.1]. Similarly, Fig. 1.6 gives the term-wise solution
plots for r  0.2 and β  0.1. The TFN fuzzy plots for a fixed y  20 and time t 
0.2, for various x values, are incorporated in Fig. 1.7.
Term-wise solutions of 2D SWWE with crisp basin depth at a fixed time t  0.5
for fixed values of x  22 and y  22 are presented in Table 1.1.

√
Fig. 1.3 Term-wise solution η plots of 2D SWWE at time t  2 s and h  2/10
18 P. Karunakar and S. Chakraverty

√
Fig. 1.4 Water surface elevation of 2D SWWE at time a t  2.5 s and b t  3 s, h  2/10

From Fig. 1.3, one may observe that the convergent solution of 2D SWWE is
obtained as the number of terms is increasing. Specially, after sixth term, there is no
change in the peak height. This indicates minimum six terms are necessary to get
a convergent solution for the present problem. The number of terms may differ for
other problems, but one may expect convergent solution as we increase the number
of terms. And from Figs. 1.3 and 1.4, it may be noted that at different time levels
t  2 s, t  2.5 s, and t  3 s, the water surface elevation η is almost same, but
wavelength (that is width) of the wave at time t  3 s is more than that of time t 
2 s. From this, it may be understood that as time increases, wavelength increases.
It is worth mentioning that this is main characteristic of shallow water waves that
is wavelength of the shallow water waves is more than that of depth of the fluid.
The convergence of HPM solution of shallow water equations with uncertain basin
depth h̃  [0.1, 0.1, 0.1] can also be clearly seen clear from Figs. 1.5 and 1.6. In this
case, two sets of particular values of the parameters r and β have been considered to
show that the solution is converging (viz. r  1 and β  0; r  0.2 and β  0.1).
Figure 1.7 clearly demonstrates the uncertain solution in term of fuzzy plots by fixing
the values of y and t (viz. y  20 and t  0.2), with TFN basin depth. Similarly, one
may generate other fuzzy uncertain solution by taking different values of y and t.
1 2-D Shallow Water Wave Equations with Fuzzy Parameters 19

Fig. 1.5 Water surface elevation of 2D SWWE with fuzzy basin depth (center solution r  1, β 
0, t  0.1 s)
Another random document with
no related content on Scribd:
 TO STEW A RUMP OF BEEF.

This joint is more easily carved, and is of better appearance when

the bones are removed before it is dressed. Roll and bind it firmly
with a fillet of tape, cover it with strong cold beef broth or gravy, and
stew it very gently indeed from six hours to between seven and
eight; add to it, after the scum has been well cleared off, one large or
two moderate-sized onions stuck with thirty cloves, a head of celery,
two carrots, two turnips, and a large faggot of savoury herbs. When
the beef is perfectly tender quite through, which may be known by
probing it with a sharp thin skewer, remove the fillets of tape, dish it
neatly, and serve it with a rich Espagnole, and a garnish of forced
tomatas, or with a highly-flavoured brown English gravy, and stewed
carrots in the dish: for these last the mild preparation of garlic or
eschalots, of page 122, may be substituted with good effect. They
should be well drained, laid round the meat, and a little brown gravy
should be poured over the whole.
This is the most simple and economical manner of stewing the
beef; but should a richer one be desired, half roast the joint, and
stew it afterwards in strong gravy to which a pint of mushrooms, and
a pint of sherry or Madeira, should be added an hour before it is
ready for table. Keep it hot while a portion of the gravy is thickened
with a well-made brown roux (see Chapter IV.), and seasoned with
salt, cayenne, and any other spice it may require. Garnish it with
large balls of forcemeat highly seasoned with minced eschalots,
rolled in egg and bread-crumbs, and fried a fine golden brown.
Plainly stewed from 6 to 7 or 8 hours. Or: half roasted then stewed
from 4 to 5 hours.
Obs.—Grated horseradish, mixed with some well-thickened brown
gravy, a teaspoonful of mustard, and a little lemon-juice or vinegar, is
a good sauce for stewed beef.
 BEEF PALATES. (ENTRÉE.)

First rub them well with salt, to cleanse them well; then wash them
thoroughly in several waters, and leave them to soak for half an hour
before they are dressed. Set them over the fire in cold water, and
boil them gently until the skin will peel off, and the palates are
tolerably tender. It is difficult to state the exact time required for this,
as some will be done in two hours and a half, and others in not less
than from four to five hours. When thus prepared, the palates may
be cut into various forms, and simmered until fit to serve in rich
brown gravy, highly flavoured with ham, cayenne, wine, and lemon-
peel; or they will make an excellent currie. As they are very insipid of
themselves, they require a sauce of some piquancy, in which, after
they have been peeled and trimmed, they should be stewed from
twenty to thirty minutes, or until they are perfectly tender. The black
parts of them must be cut away, when the skin is taken off. An onion,
stuck with a few cloves, a carrot sliced, a teaspoonful of whole white
pepper, a slice of butter, and a teaspoonful of salt, may be boiled
with the palates in the first instance; and they will be found very
good, if sent to table in the curried gravy of Chapter XVI., or in the
Soubise of Chapter VI., made thinner than the receipts direct.
Boiled from 2-1/2 to 4 or 5 hours. Stewed from 20 to 30 minutes.
Obs.—A French cook of some celebrity, orders the palates to be
laid on the gridiron until the skin can be easily peeled or scraped off;
the plan seems a good one, but we have not tried it.
 BEEF PALATES.

(Neapolitan Mode.)
Boil the palates until the skin can be easily removed, then stew
them very tender in good veal broth, lay them on a drainer and let
them cool; cut them across obliquely into strips of about a quarter-
inch in width, and finish them by either of the receipts for dressing
maccaroni, which will be found in Chapters XVIII. and XX.
 STEWED OX-TAILS.

They should be sent from the butcher ready jointed. Soak and
wash them well, cut them into joints or into lengths of two or three
joints, and cover them with cold broth or water. As soon as they boil
remove the scum, and add a half-teaspoonful of salt or as much
more as may be needed, and a little common pepper or cayenne, an
onion stuck with half a dozen cloves, two or three small carrots, and
a branch or two of parsley. When these have simmered for two hours
and a quarter, try the meat with a fork, and should it not be perfectly
tender, let it remain over the fire until it is so. Ox-tails sometimes
require nearly or quite three hours’ stewing: they may be served with
the vegetables, or with the gravy strained from them, and thickened
like the English stew of the present chapter.
Ox-tails, 2; water or broth to cover them; salt, 1/2 teaspoonful, or
more; little pepper or cayenne; onion, 1; cloves, 6; carrots, 2 or 3;
parsley, 2 or 3 branches: 2-1/2 to 3 hours.
 BROILED OX-TAIL. (ENTRÉE.)

(Very good.)
When the ox-tail is ready for the stewpan, throw it into plenty of
boiling water slightly salted, and simmer it for fifteen minutes; then
take it up and put it into fresh water to cool; wipe it, and lay it round
in a small stewpan without dividing it, just cover it with good beef
gravy, and stew it gently until very tender; drain it a little, sprinkle
over it a small quantity of salt and cayenne, dip it into clarified butter
and then into some fine bread-crumbs, with which it should be thickly
covered, lay it on the gridiron, and when equally browned all over
serve it immediately. If more convenient the ox-tail may be set into
the oven or before the fire, until properly coloured: it may likewise be
sent to table without broiling, dished upon stewed cabbage or in its
own gravy thickened, and with tomata sauce, in a tureen.
 TO SALT AND PICKLE BEEF, IN VARIOUS WAYS.

Let the meat hang a couple of days in mild weather, and four or
five in winter, before it is salted or pickled. During the heat of
summer it is better to immerse it entirely in brine, that it may be
secured alike from the flies, and from the danger of becoming putrid.
Trim it, and take out the kernels from the fat; then rub a little fine dry
salt over it, and leave it until the following day; drain it well from the
blood, which will be found to have flowed from it, and it will be ready
for any of the following modes of curing, which are all excellent of
their kind, and have been well proved.
In very cold weather, the salt may be applied quite warm to the
meat: it should always be perfectly dry, and reduced to powder.
Saltpetre hardens and renders the meat indigestible; sugar, on the
contrary, mellows and improves it much; and it is more tender when
cured with bay salt than when common salt is used for it.
 TO SALT AND BOIL A ROUND OF BEEF.

Mix an ounce of saltpetre, finely powdered, with half a pound of

very coarse sugar, and rub the beef thoroughly with them; in two
days add three quarters of a pound of common salt, well dried and
beaten; turn and rub the meat well in every part with the pickle for
three weeks, when it will be fit to dress. Just wash off the salt, and
skewer the beef as round and as even as possible; bind it tightly with
broad tape, cover it with cold water, place it over a rather brisk fire,
and after it boils draw it to the side of the stove and let it simmer
gently for at least five hours. Carrots, mashed turnips, or cabbages,
are usually served with boiled beef; and horseradish stewed for ten
minutes in equal parts of vinegar and water, then pressed well from
them, and mixed with some rich melted butter, is a good sauce for it.
Beef, 20 lbs.; coarse sugar, 1/2 lb.; saltpetre, 1 oz.: 2 days. Salt,
3/4 lb.: 21 days. Boil 5 hours, or more.
Obs.—Beef cured by this receipt if properly boiled, is tender, of
good colour and flavour, and not over salt. The rump, edgebone, and
brisket may be salted, or pickled in the same way as the round.
 HAMBURGH PICKLE FOR BEEF, HAMS, AND TONGUE.

Boil together, for twenty minutes, two gallons of water, three

pounds of bay salt, two pounds of coarse sugar, two ounces of
saltpetre, and two of black pepper, bruised, and tied in a fold of
muslin; clear off the scum thoroughly, as it rises, pour the pickle into
a deep earthen pan, and when it is quite cold lay in the meat, of
which every part must be perfectly covered with it. A moderate-sized
round of beef will be ready for table in a fortnight; it should be turned
occasionally in the brine. Five pounds of common salt may be
substituted for the quantity of bay salt given above; but the meat will
not be so finely flavoured.
Water, 2 gallons; bay-salt, 3 lbs.; saltpetre, 2 oz.; black pepper, 2
oz.; sugar, 2 lbs.: 20 minutes.
 ANOTHER PICKLE FOR TONGUES, BEEF, AND HAMS.

To three gallons of spring water add six pounds of common salt,

two pounds of bay-salt, two pounds of common loaf sugar, and two
ounces of saltpetre. Boil these over a gentle fire, and be careful to
take off all the scum as it rises: when quite cold it will be fit for use.
Rub the meat to be cured, with fine salt, and let it drain for a day in
order to free it from the blood; then immerse it in the brine, taking
care that every part of it shall be covered. Young pork should not
remain more than from three to five days in the pickle; but hams for
drying may be left in it for a fortnight at least; tongues will be ready in
rather less time. Beef may remain from one week to two, according
to its size, and the degree of saltness desired for it. A little
experience will soon teach the exact time required for the different
kinds of meat. When the pickle has been in use for about three
months, boil it up again gently, and take the scum carefully off. Add
to it three pounds of common salt, four ounces of sugar, and one of
saltpetre: it will remain good for many months.
Water, 3 gallons; common salt, 6 lbs.; bay salt, 2 lbs.; loaf sugar, 2
lbs.; saltpetre, 2 oz.: boil 20 to 30 minutes.
 DUTCH, OR HUNG BEEF.

For fourteen pounds weight of the round, the rump, or the thick
flank of beef, mix two ounces of saltpetre with the same quantity of
coarse sugar; rub the meat with them in every part, and let it remain
for two days, then add one pound of bay salt, four ounces of
common salt, and one ounce of ground black pepper. Rub these
ingredients thoroughly into the beef, and in four days pour over it a
pound of treacle; rub and turn it daily for a fortnight; drain, and send
it to be smoked. When wanted for table, put it into plenty of boiling
water, boil it slowly, and press it under a heavy weight while hot. A
slice of this beef, from which the edges have been carefully trimmed,
will serve to flavour soups or gravies as well as ham.
Beef, 14 lbs.; saltpetre and coarse sugar, each 2 oz.: 2 days. Bay
salt, 1 lb.; common salt, 4 oz.: pepper, 1 oz.: 4 days. Treacle, 1 lb.:
14 days.
Obs.—Three quarters of a pound of coarse sugar may be rubbed
into the meat at first, and the treacle may be altogether omitted;
cloves and mace, too, may be added in the same proportion as for
spiced beef.
 COLLARED BEEF.

Only the thinnest part of the flank, or the ribs, which are not so
generally used for it, will serve conveniently for collaring. The first of
these should be hung in a damp place for a day or two, to soften the
outer skin; then rubbed with coarse sugar, and left for a couple of
days; when, for eight pounds of the meat, one ounce of saltpetre and
half a pound of salt should be added. In ten days it will be fit to
dress. The bones and tough inner skin must be removed, and the
beef sprinkled thickly on the under side with parsley and other
savoury herbs shred small, before it is rolled, which should be done
very tightly: it must then be secured with a cloth, and bound as
closely as possible with broad tape. It will require nearly or quite five
hours of gentle boiling, and should be placed while hot under a
weight, or in a press, without having the tape and cloth removed.
Beef, 8 lbs.; sugar, 3 oz.; salt, 8 oz.: 10 days. Boil 5 hours.
 COLLARED BEEF.

(Another way.)

Mix half an ounce of saltpetre with the same quantity of pepper,

four ounces of bay salt, and four of common salt; with these rub well
from six to seven pounds of the thin flank, and in four days add
seven ounces of treacle; turn the beef daily in the pickle for a week
or more; dip it into water, bone it and skin the inside, roll and bind it
up very tightly, lay it into cold water, and boil it for three hours and a
half. We have found beef dressed by this receipt extremely good:
herbs can, of course, be added to it as usual. Spices and juniper
berries would to many tastes improve it, but we give the receipt
simply as we have been accustomed to have it used.
Thin flank, 6 to 7 lbs.; bay salt, and common salt, each 4 oz.;
saltpetre, 1/2 oz.; pepper, 1/2 oz.: 4 days. Treacle, 7 oz.: 8 to 10
days. Boiled 3-1/2 hours.
 A COMMON RECEIPT FOR SALTING BEEF.

One ounce of saltpetre, and a pound of common salt, will be

sufficient for sixteen pounds of beef. Both should be well dried, and
finely powdered; the saltpetre rubbed first equally over the meat, and
the salt next applied in every part. It should be rubbed thoroughly
with the pickle and turned daily, from a week to ten days. An ounce
or two of sugar mixed with the saltpetre will render the beef more
tender and palatable.
Beef, 16 lbs.; saltpetre, 1 oz.; salt, 1 lb.: 7 to 10 days.
 SPICED ROUND OF BEEF.

(Very highly flavoured.)

Rub the beef well in every part with half a pound of coarse brown
sugar, and let it remain two days; then reduce to powder, and mix
thoroughly before they are applied to the meat, two ounces of
saltpetre, three quarters of a pound of common salt, a quarter of a
pound of black pepper, three ounces of allspice, and four of bruised
juniper-berries. Rub these ingredients strongly and equally over the
joint, and do so daily for three weeks, turning it at the same time.
Just wash off the spice, and put the beef into a tin, or covered
earthen pan as nearly of its size as possible, with a cup of water or
gravy; cover the top thickly with chopped beef-suet, and lay a coarse
thick crust over the pan; place the cover on it, and bake the meat
from five to six hours in a moderate oven, which should not,
however, be sufficiently fierce to harden the outside of the joint,
which, if properly managed will be exceedingly tender. Let it cool in
the pan; and clear off the suet before it is dished. It is to be served
cold, and will remain good for a fortnight.
Beef, 20 to 25 lbs. weight; sugar, 3 oz.: 2 days. Saltpetre, 2 oz.;
common salt, 3/4 lb.; black pepper, 4 oz.; allspice, 3 oz.; juniper-
berries, 4 oz.: 21 days. Baked 5 to 6 hours.
Obs.—We have not ourselves tested this receipt, but the meat
cured by it has received such high commendation from several of
our friends who have partaken of it frequently, that we think we may
safely insert it without. The proportion of allspice appears to us more
than would be agreeable to many tastes, and we would rather
recommend that part of it should be omitted, and that a portion of
nutmeg, mace, and cloves, should be substituted for it; as we have
found these spices to answer well in the following receipt.
 SPICED BEEF.

(Good and wholesome.)

For twelve pounds of the round, rump, or thick flank of beef, take a
large teaspoonful of freshly-pounded mace, and of ground black
pepper, twice as much of cloves, one small nutmeg, and a quarter of
a teaspoonful of cayenne, all in the finest powder. Mix them well with
seven ounces of brown sugar, rub the beef with them and let it lie
three days; add to it then half a pound of fine salt, and rub and turn it
once in twenty-four hours for twelve days. Just wash, but do not
soak it; skewer, or bind it into good form, put it into a stewpan or
saucepan nearly of its size, pour to it a pint and a half of good beef
broth, and when it begins to boil, take off the scum, and throw in one
small onion, a moderate-sized faggot of thyme and parsley, and two
large, or four small carrots. Let it simmer quite softly for four hours
and a half, and if not wanted to serve hot, leave it in its own liquor
until it is nearly cold. This is an excellent and far more wholesome
dish than the hard, bright-coloured beef which is cured with large
quantities of salt and saltpetre: two or three ounces of juniper-berries
may be added to it with the spice, to heighten its flavour.
Beef, 12 lbs.; sugar, 7 oz.; mace and black pepper, each, 1 large
teaspoonful; cloves, in powder, 1 large dessertspoonful; nutmeg, 1;
cayenne, 1/4 teaspoonful: 3 days. Fine salt, 1/2 lb.: 12 days. Beef
broth (or bouillon), 1-1/2 pint; onion, 1 small; bunch of herbs; carrots,
2 large, or 4 small: stewed 4-1/2 hours.
Obs.—We give this receipt exactly as we have often had it used,
but celery and turnips might be added to the gravy; and when the
appearance of the meat is much considered, three-quarters of an
ounce of saltpetre may be mixed with the spices; the beef may also
be plainly boiled in water only, with a few vegetables, or baked in a
deep pan with a little gravy. No meat must ever be left to cool in the
stewpan or saucepan in which it is cooked; it must be lifted into a
pan of its own depth, and the liquor poured upon it.
 A MINIATURE ROUND OF BEEF.

“Select a fine rib of beef, and have it cut small or large in width
according to your taste; it may thus be made to weigh from five to
twelve pounds, or more. Take out the bone, and wrap the meat
round like a fillet of veal, securing it with two or three wooden
skewers; place it in a strong pickle for four or five days, and then
cook it, taking care that it does not boil, but only simmers, from forty
minutes, or more, according to its size. It is best to put it on in hot
water, as it will not draw the gravy so much as cold. Many persons
adjust a rib of beef in this way for roasting: let them try it salted, and
they need not envy the possessor of the finest round of beef.” We
give the receipt to our readers in its original form, and we can assure
them, from our own experience, that it is a good one; but we would
recommend that, in dressing the meat, quite the usual time for each
pound of it should be allowed. When boned and rolled at the
butcher’s, the skewers should be removed when it is first brought in;
it should be well wiped with a dry cloth, or washed with a little fresh
brine, and a small quantity of salt and saltpetre should be rubbed
over the inside, it may then be firmly bound with tape, and will be
quite ready to boil when taken from the pickle. The sirloin, after the
inside fillet is removed, may be cured and dressed in the same way,
and will be found super-excellent if the beef be well fatted and
properly kept. The Hamburgh pickle (see page 197) is perhaps the
best for these joints. Part of the rump, taken clear of bone, answers
admirably when prepared by this receipt.
 BEEF ROLL, OR CANELLON DE BŒUF. (ENTRÉE.)

Chop and mix thoroughly two pounds of lean and very tender beef
with one pound of slightly striped bacon; season them with a large
teaspoonful of pepper, a little salt, a small nutmeg, or two-thirds as
much mace, the grated rind of a lemon, or a teaspoonful of thyme
and parsley finely minced. Form the whole into a thick rouleau, wrap
a buttered paper round it, enclose it in a paste made of flour and
water, and send it to a moderate oven for a couple of hours. Remove
the paper and the crust, and serve the meat with a little brown gravy.
Lamb and veal are excellent dressed in this way, particularly when
mixed with plenty of mushrooms. Brown cucumber sauce should be
served with the lamb; and currie, or oyster sauce, when there are no
mushrooms, with the veal. A flavouring of onion or of eschalot,
where it is liked, can be added at pleasure to the beef: suet, or the
fat of the meat, may be substituted for the bacon.
Beef, 2 lbs.; bacon, 1 lb.; pepper, 1/4 oz.; little salt; small nutmeg;
rind of 1 lemon, or savoury herbs, 1 tablespoonful: baked 2 hours.
 MINCED COLLOPS AU NATUREL.

Mince finely a pound of very tender rump steak, free from fat or
skin; season it with a moderate quantity of pepper and salt, set it
over a gentle fire, and keep it stirred with a fork until it is quite hot
that it may not gather into lumps. Simmer it very slowly in its own
gravy from ten to twelve minutes, and then, should it be too dry, add
a little boiling water, broth, or gravy; stew it for two minutes longer,
and serve it directly.
These collops are particularly suited to persons in delicate health,
or of weak digestion; and when an extra dish is required at a short
notice, from the expedition with which they may be dressed, they are
a convenient resource.
10 to 12 minutes.
 SAVOURY MINCED COLLOPS.

Make a little thickening (see Brown Roux, Chapter V.) with about
an ounce and a half of butter, and a dessertspoonful of flour; when it
begins to be coloured, shake lightly into it a large teaspoonful of
finely-shred parsley or mixed savoury herbs, two-thirds as much of
salt, and half the quantity of pepper. Keep these stirred over a gentle
fire until the thickening is of a deep yellow brown; then add a pound
of rump-steak, finely minced, and keep it well separated with a fork
until it is quite hot; next pour to it gradually half a cupful of boiling
water, and stew the collops very gently for ten minutes. Before they
are served, stir to them a little catsup, chili vinegar or lemon-juice: a
small quantity of minced onion, eschalot, or a particle of garlic, may
be added at first to the thickening when the flavour is not objected to.
 A RICHER VARIETY OF MINCED COLLOPS.

Omit the minced herbs from the thickening, and season it with
cayenne and a small quarter of a teaspoonful of pounded mace.
Substitute beef gravy for the boiling water, and when the collops are
nearly done, fill a wineglass with one fourth of mushroom catsup,
and three of port wine, and stir these to the meat. Serve the collops
very hot, and garnish them with alternate forcemeat balls (see No. 1,
Chapter VIII.) and fried sippets. If flavoured with a little gravy made
from the bones of a roast hare, and served with currant jelly, these
collops will scarcely be distinguished from game.