Page 1
* Corresponding Author. Email address: [email protected] , Tel: +60175551703.
The Variational Iteration Method for Solving
Fuzzy Duffing's Equation
A.F. Jameel*
School of Mathematical Sciences, 11800 USM, University Science Malaysia
Copyright 2014 © A.F. Jameel. This is an open access article distributed under the Creative Commons Attribution
License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is
properly cited.
Abstract
In this paper, we use the Variational Iteration Method (VIM) to find the approximate analytical solution
for an initial value problem involving the fuzzy Duffing ordinary differential equation. VIM allows for the
solution of the differential equation to be calculated in the form of an infinite series in which the
components can be easily computed. Numerical examples were solved to illustrate the capability of VIM
in this regard.
Keywords: Fuzzy numbers, Fuzzy differential equations, Variational Iteration Method and Duffing Equation.
1 Introduction
Nonlinear ordinary differential equations are used to model a wide class of problems. The dynamics of
these equations can often be understood by reference to a simple nonlinear ordinary differential equation.
One such equation is the second order Duffing equation given by [9]
( ) ( ) ( ) ( ) ( ) (1.1)
The initial conditions associated with it are
( ) ( )
where , , , , and are real constants. Equation (1.1) has been discussed in many works that arise in
various scientific fields such as physics, engineering, biology, and communication theory [9, 24, 28, 35].
Various problems involving nonlinear control systems in physics, engineering and communication theory
have increased interest in the periodical forced Duffing's equation [23, 25, 28]. In some cases, uncertainty
leads to the fuzzy Duffing’s equation (FDE). These models are used in various applications including
population models [36], quantum optics gravity [13], and medicine [6, 8]. Fuzzy Initial value problem
Journal of Interpolation and Approximation in Scientific Computing 2014 (2014) 1-14
Available online at www.ispacs.com/jiasc
Volume: 2014, Year 2014 Article ID: jiasc-00053, 14 Pages
doi:10.5899/2014/jiasc-00053
Research Article
Page 2
of 142Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
(FIVP) involving ordinary differential equations are suitable mathematical models to model certain
dynamical systems in which there exist uncertainties or vagueness.
In recent years approximate-analytical methods such as Adomian Decomposition Method (ADM),
Homotopy Perturbation Method (HPM) and Variational Iteration Method (VIM) have been used to solve
fuzzy initial value problems which involve ordinary differential equations. In [16], the HPM was used to
solve first order linear fuzzy initial value problems involving ordinary differential equations. The ADM
was employed in [2, 7, 17] to solve first order linear and nonlinear fuzzy initial value problems. VIM is an
approximate - analytical method that was first proposed by He [18-22] and has been applied to many
physics and engineering problems [5, 27, 31, 34]. VIM was used in [17] to obtain the approximate solution
for first order linear FIVP. Furthermore, it was found that VIM is more effective than ADM and the
convergence of VIM is much faster than ADM. In [1] VIM also was used to find the approximate-
analytical solution for fuzzy differential equations including nonlinear first order FIVP. Abbasbandy et al.
in [3] used VIM to find the approximate solution for high order linear FIVP by converting it into a first
order system of fuzzy differential equations. The convergence of VIM for this system was also proved.
In this paper, we use VIM to obtain an approximate-analytical solution of nonlinear FIVP involving fuzzy
Duffing’s equation directly without converting into a first order system. The convergence of VIM is also
investigated. The outline of this paper is as follows. In section 2, some basic definitions and notations are
given as they will be used in other sections. In section 3, the structure of VIM is formulated for solving
second order FIVP. In section 4, the convergence analysis of VIM is presented and proved in detail. In
section 5, we consider some examples and finally, in section 6, we give the conclusions of this study.
2 Preliminaries
The definitions and theorems given in this section are required in our work. We cite articles from where
they were substantively taken
Definition 2.1. A fuzzy number is a function such as , - which satisfies the following
properties[17]
1. ( ) is normal, i.e with ( ) ,
2. ( ) is convex fuzzy set ,i.e. ( ( ) ) * ( ) ( )+ , -,
3. upper semi-continuous on ,
4. * ( ) + is compact.
A fuzzy number can also be represented in the parametric form
Definition 2.2. In the parametric form, a fuzzy number is represented by an ordered pair of
functions ( ( ) ( )) , , - which satisfies [24]:
i. ( ) is a bounded left continuous non-decreasing function over , -.
ii. ( ) is a bounded left continuous non-increasing function over , -.
iii. ( ) ( ), , -
A crisp number r is simply represented by ( ) ( ) , , -
Definition 2.3. If be the set of all fuzzy numbers, we say that ( ) is a fuzzy function [17] if
Page 3
of 143Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
Definition 2.4. A mapping for some interval is called a fuzzy function process [14] and
we denote the r-level set by:
, ( )- 0 ( ) ( )1 , -
The r-level sets of a fuzzy number are much more effective as representation forms of fuzzy set. Fuzzy sets
can be defined by the families of their r-level sets based on the resolution, identity theorem [12, 33].
Definition 2.5. The fuzzy integral of fuzzy process [31], ( ) ∫ ( )
for and , - is defined by:
∫ ( )
6∫ ( )
∫ ( )
7
Definition 2.6. Each function induces another function ( ) ( ) defined for each fuzzy
interval in by[37]:
( )( ) 8 ( ) ( ) ( )
( )
This is called the Zadeh extension principle.
Definition 2.7. Consider . If there exists such that , then is called the H-
difference[30] (Hukuhara difference) of x and y and is denoted by .
Definition 2.8. [31] Let ( ) be the Hausdorff distance between two fuzzy sets ,
( ) * (, - , - )| , -+
and ( ) is a complete metric space.
Definition 2.9. Let and , where , -. We say that Hukuhara differentiable at if
there exists an element [ ]
such that for all h > 0 sufficiently near to 0, ( )
( ) ( ) ( ) exists where the limits are taken in the metric space ( )[31]
( ) ( )
( ) ( )
The fuzzy set ( ) is called Hukuhara derivative of at .
Theorem 2.1. [31] Let , - be Hukuhara differentiable and denote
[ ( )]
0 ( ) ( )1 .
Then the boundary functions 0 ( ) ( )1 are differentiable functions and
[ ( )]
[( ( ))
. ( )/ ] , , -
Page 4
of 144Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
Theorem 2.2. [31] Let , - be Hukuhara differentiable n times and
[ ( )]
0 ( ) ( )1
0 ( ) ( )1 , -
that ( ) ( ) are differentiable. We can write for nth order fuzzy derivative
[ ( )( )]
6( ( )( ))
4 ( )
( )5
7 , - .
3 Fuzzification and Defuzzification of VIM for Second Order FIVP
The general structures of VIM for solving crisp nonlinear problems have been presented in [18-22]
amongst others. The general defuzzification of nth order FIVP was presented in [31]. The basic principles
of VIM for FIVP have been described in [1, 17, 31]. In this section the structure of VIM for the
approximate solution of second order FIVP is described followings the work of others. Consider the
following second order FIVP:
( ) ( ( ) ( )) ( ) , - (3.2)
( ) ( )
(3.3)
Here ( ) is the fuzzy function of the crisp variable with being a fuzzy function of the crisp variable,
fuzzy variable and the first order fuzzy Hukuhara derivative ( ). According to theorem 2.2 section 2
( ) is the second order fuzzy Hukuhara derivative of the fuzzy function with are fuzzy
numbers as in definitions (2.1-2.2) section 2. We denote the fuzzy function by , - , - , for
, - and , -. It means that the r-level set of ( ) can be defined as:
{, ( )- 0 ( ) ( )1
, ( )- 0 ( ) ( )1
(3.4)
and
{, ( )- 0 ( ) ( )1
, ( )- 0 ( ) ( )1
(3.5)
Now for defuzzification of Eq. (3.2), we let ( ) ( ) ( ) be the fuzzy function of ( ) ( ), such
that
( ) 0 ( ) ( )1 0 ( ) ( ) ( ) ( )1
Also, we can write
[ ( )]
0 ( ) ( )1 (3.6)
with
{ ( ) 0 1
( ) 0 1
(3.7)
Page 5
of 145Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
Since ( ) ( ( )) and by using the fuzzy extension principle as in section 2, we can define the
following membership function
{ ( ( ) ) { ( ( ))| ( ) ( )}
( ( ) ) { ( ( ))| ( ) ( )} (3.8)
where
{ ( ( ) ) ( ( ) ( ))
( ( ) ) ( ( ) ( )) (3.9)
Now we can rewrite Eq. (1.1) as follows
( ) ( ( ) ( )) ( ) (3.10)
( ) ( ( ) ( )) ( ) (3.11)
According to the VIM procedure, we can construct the following correction functional as follows
( ) ( ) ∫ ( ) 2 ( ) . ( )
⏞ / ( )3
(3.12)
( ) ( ) ∫ ( ) 2
( ) . ( )
⏞ / ( )3
(3.13)
where , , -, ( ) is a general Lagrange multiplier. Now we let
⏞ . ( )
⏞ / ( ) (3.14)
⏞ . ( )
⏞ / ( ) (3.15)
where and are nonlinear operators , then we can rewrite Esq. (3.14) and (3.15) in the following forms
( ) ( ) ∫ ( ) 2 ( )
⏞3
(3.16)
( ) ( ) ∫ ( ) 2
( )
⏞3
(3.17)
⏞ is considered as restricted variation i.e
⏞ . The general Lagrangian multiplier ( ) related
with Eq. (3.2) can be determined as:
( ) ( ) ∫ ( ) 2 ( )
⏞3
( ) ( ) ∫ ( ) 2 ( )3
Integrating by parts we obtain the following:
( ) , ( )- ( ) ( ) ( ) ∫ ( ) ( )
Therefore, have the following stationary conditions:
{
( )
( ) |
( )|
(3.18)
Page 6
of 146Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
Similarly, we have the same Lagrangian multiplier for the upper bound of Eq. (3.2). From these conditions
and according to the order of the Eq. (3.16) we can determine the Lagrangian multiplier
( ) ( ) (3.19)
Here the initial guesses that satisfies the initial conditions in Eq. (3.3) are given by
{ ( ) ( ) ( )
( ) ( ) ( )
(3.20)
where ( ) ( ) are fuzzy constants numbers as in definition (2.1) of section 2 and can be determined
from the initial conditions in Eq. (3.3) for all , -.
The successive approximations in Esq. (3.12)-(3.13) of VIM will be readily obtained by choosing all the
above-mentioned parameters ( ) and ( ). Consequently, for , the exact solution may
be obtained by
( ) ( ) *, - + (3.21)
where ( ) [ ( ) ( )] is the exact solution of Eq. (3.2) and the VIM approximate solution is
( ) 0 ( ) ( )1. We reiterate that basic principles of VIM for FIVP have been described in
the work of previous researchers as cited in this section.
4 Convergence of VIM
In [3] the convergence of the approximate solution of system of linear FIVP obtained by VIM have been
proved. Also the proof of convergence of VIM for second order crisp nonlinear differential equation was
presented in [15]. Using these previous works as a basis, we investigate the convergence of VIM for a
nonlinear FIVP such as fuzzy Duffing equation. The variational iteration formula generates a recurring
sequence *, - +. Clearly, the limit of the sequence will be the solution of Eq. (3.2) if the sequence is
convergent. In this section, we give a proof of convergence of VIM by introducing a new iterative
formulation of this procedure. The proof follows the procedure used by and ideas of earlier researchers.
Here, , - denotes the class of all real valued functions defined on , - which have continuous
second order derivatives. Before we continue the discussion of convergence details we note than one can
rewrite equations (14-15) in the following form
, ( )- , ( )- (4.22)
where the linear operator
can be defined as
, ( )- 8 ( )
( )
(4.23)
and the nonlinear operator can be defined as
, ( )- { ( . ( )/ ( ))
. . ( )/ ( ) / (4.24)
Lemma 4.1. If for any i, , - , -, then the variational iteration formula (12-13) is equivalent to
the following iterative relation for all , -:
( ( ) ( )) ( ( ) ( )
( )) (4.25)
.
( ) ( )/ .
( )
( )
( )/ (4.26)
Page 7
of 147Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
where
{ ( ( )
( ) ( )) ( ) . ( )/ ( )
. ( )
( )
( )/
( ) . ( )/ ( )
(4.27)
and as in Eq.(4.22) .
Proof. First we take the lower bound solution of Eq. (4.22). Suppose ( ) ( ) satisfies the
variational iteration formula (3.12). Applying to both sides of (3.12) we have
( ( ) ( )) ∫
( )
( )
|
[ ( )| ]
Now, by using the conditions (3.18) and ( )
|
, we obtain
( ( ) ( )) ( ( )
( ) ( ))
Conversely, suppose ( ) ( ) satisfies (4.25). From the definition of in (4.22) and
( ) , multiplying (4.24) by ( ) and then integrating from both sides of the resulting term from 0
to t we obtain
∫ ( ) ( ( )
( ))
∫ ( )
(4.28)
Integration Eq. (4.28) by parts, then Eq. (4.25) becomes
( )| ( ( )
( )) ( )
|
( ( ) ( )) ∫ ( )
( ( )
( )) ∫ ( )
which exact results in Eq. (3.12) upon imposing the conditions (3.18). The upper bound of Eq. (3.2) can be
proved in a similar manner.
Theorem 4.1. If the sequence (3.21) converges, where *, - + is produced by the VIM formulation of (12-
13), then it must be the exact solution of the Eq. (3.2) for all , -.
Proof. If the sequence *, - + converges, we can write
8 ( ) ( )
( ) ( )
(4.29)
and it holds for
8 ( ) ( )
( )
( ) (4.30)
Now for the lower bound of Eq. (3.2), we use Eq. (4.29), and the definition of in (4.23), we can obtain
( ( ) ( )) ( ( ) ( )) (4.31)
From Eq. (4.31) and according to the Lemma 4.1, we can obtain
Page 8
of 148Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
( ( ) ( )) ( ( ) ( )
( )) (4.32)
which gives us
( ( ) ( )
( )) (4.33)
Since . ( )/ ( ( ) ( )) and Eq. (3.2) is a second order equation and if ,
may be include both ( ) ( ) or one of them such that for the lower bound
( ( )) ( ) , ( ( )) (
( ))
( )
Now from (4.32) and (4.26), one can derive the following
( ( ) ( )
( )) ( ( ) . ( )/ ( )) (4.34)
(
( ))
( . ( )/) . ( )/
. ( ) ( ( ) ( )) ( )/
From the equations (4.33) and (4.34),
( ) ( ( ) ( )) ( ) for (4.35)
On the other hand, using the specified initial conditions in (3.3) and the initial guess in (3.20), we have
{ ( )
( ) ( )
( )
( ) ( )
(4.36)
Therefore, according to expressions (4.33)-(4.35), ( ) must be the exact solution of the Eq. (3.2) , and
similarly we work on the upper bound of Eq. (3.2). This ends the proof.
It is clear that the convergence of the sequence (3.21) depends upon the initial guess in (3.20) and the
linear operator with the number of terms of sequence (3.21). Fortunately, as long as ( ) and are
chosen that the sequence (3.21) converges in a region , it must converge to the exact solution in
this region.
5 Numerical Examples
In order to assess the advantages and the accuracy of VIM for solving fuzzy Duffing differential
equations, we have applied the method to two different examples. All the results are calculated by using
the Mathematica codes.
Example 5.1. Consider the fuzzy Duffing’s equation [4] of the following type
( ) ( ) ( ) (5.37)
( ) , - ( ) , -
, -
The exact solution of Eq. (5.37) was given in [4] in which it was solved by the Laplace transform
decomposition algorithm. The initial approximation guesses of Eq. (5.37) are given by:
( ) and ( ) ( )
Page 9
of 149Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
According to section 3 the variational formula of this problem is given as follows:
( ) ( ) ∫ ( ){ ( )( ) ( ) , ( )- }
(5.38)
In order to solve this problem and get a convergent solution we expand in Maclaurin
series expansions of order 6 and substitute these series in Eq. (5.38) such that
( )
and ( )
We use formula (5.38) to obtain 5-order VIM approximate solution ( ) results and compare with the
exact ( ) and Maple numerical solutions ( ) of Eq. (5.37) when for all , - as
shown in the following tables and figure:
Table 1: Lower 5- order VIM approximate, exact and Maples numerical solution of Eq. (5.37) at t = 0.1
Table 2: Upper 5- order VIM approximate, exact and Maples numerical solution of Eq. (5.37) at t = 0.1
( ) ( ) ( )
Fig 1: Approximate 5-order VIM compares with the exact and Maple numerical solutions of Eq. (5.37) when t =0.1.
Table 2: Upper 5- order VIM approximate, exact and Maple numerical solutions of
Eq. (37) at t =0.1.
r (0.1; ) (0.1, ) (0.1, )
0 1.1965247538384560 1.19652475400 1.1965247998087429200
0.25 0.9185228031915503 0.91852280370 0.9185229272108376140
0.5 0.6437715602672260 0.64377156090 0.6437716048754136280
0.75 0.3712203400592153 0.37122034070 0.3712203458632439840
1 0.0998334160144086 0.09983341664 0.0998334151895059942
r-level
Table 1: Lower 5- order VIM approximate, exact and Maples numerical solutions of
Eq. (37) at t =0.1
r (0.1; ) (0.1, ) (0.1, )
0 0.99465487450978190 0.9946548739 0.994654868827032580
0.25 0.71762190762759080 0.7176219071 0.7176219140579600520
0.5 0.44355788058277457 0.4435578799 0.4435578163872099240
0.75 0.17141703887020315 0.1714170382 0.1714169861319961300
1 0.099833416014408650 0.09983341664 0.09983341518950599420
Page 10
of 1401Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
From figure (1) and tables (1)-(2) one can see that the exact and the numerical results satisfies the
concepts of the fuzzy numbers in section 2. The following tables compare the accuracy of 5- order VIM
approximate solution and Maple numerical solution with the exact solution of Eq. (5.37) when t = 0.1 for
all 0 ≤ r ≤1 such that
, - [ ]
| ( ) ( )| [ ]
[ ]
| ( ) ( )|
Table 3: Absolute error for the lower and upper bound approximate solution of Eq. (5.37) at t= 0.1 by 5- order VIM
exact and Maple numerical solutions
It is clear from table 3 that the 5- order VIM approximate- analytical solution is more accurate than
results from the Maple numerical solution when compared of with exact the analytical solution of Eq.
(4.36) when for all , -.
Example 5.2. Consider the Duffing’s equation [9] with the fuzzier version of the following type
( ) ( ) ( ) (5.39)
( ) , - ( ) , -
, -
The initial approximation guesses of Eq. (5.39) are given by:
( ) ( ) and ( ) ( )
According to section 3 the variational formula of this problem is given as follows:
( ) ( ) ∫ ( ){ ( )( )
( ) , ( )- }
(5.40)
Since Eq. (5.39) does not have an exact analytical solution and in order to show the accuracy of 5- terms
VIM approximate solution we define the following residual error [23]
( ) | ( ) ( ) , ( )- |
We use formula (5.40) to obtain 5-order VIM approximate solution of Eq. (5.39) when for all
, - as shown in the table below:
Table 3: Absolute error for the lower and upper bound approximate solution of
Eq. (37) at t =0.1 by 5-order VIM exact and Maple numerical solutions
r (0.1; ) (0.1, ) (0.1, ) (0.1, )
0 6.09781 × 10 10 5.07296 × 10 9 1.61543 × 10 10 4.58087 × 10 8
0.25 5.27591 × 10 10 6.95796 × 10 9 5.08449 × 10 10 1.23511 × 10 7
0.5 6.82774 × 10 10 6.35127 × 10 8 6.32773 × 10 10 4.39754 × 10 8
0.75 6.70203 × 10 10 5.20680 × 10 8 6.40784 × 10 10 5.16324 × 10 9
1 6.25591 × 10 10 1.450494 × 10 9 6.25591 × 10 10 1.450494 × 10 9
Page 11
of 1400Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
Table 4: Result and accuracy for the lower and upper bound approximate solution of
Eq. (5.39) at t= 0.1 by 5- order VIM
It is clear from table 4 that the 5-terms of VIM approximate solution of Eq. (5.39) satisfy the concepts of
the fuzzy numbers in section 2.
6 Conclusions
In this paper, the Variational Iteration Method (VIM) has been successfully introduced and applied to
solve second order fuzzy Duffing’s equation to obtain an approximate solution. The problem was solved
directly without it first being reduced to a first order system. The convergence analyses of VIM have been
investigated in detail with regard to the fuzzy Duffing’s equation following the procedures and ideas of the
previous researchers. The obtained results in this paper show that the VIM is capable of accurately solving
second order nonlinear fuzzy initial value problems.
Acknowledgement
Many thanks to Professor Ahmad Izani Md Ismail for his valuable remarks and comments which have
substantially improved the presentation of this paper.
References
[1] T. Allahviranloo, A. Panahi, H. Rouhparvar, A Computational Method To Find An Approximate
Analytical Solution For Fuzzy Differential Equations, An. S,t. Univ. Ovidius Constant, 17 (1) (2009)
5-14.
[2] T. Allahviranloo, S. Khezerloo, M. Mohammadzaki, Numerical Solution for Differential Inclusion by
Adomian Decomposition Method, Journal of Applied Mathematics, 5 (17) (2008) 51-62.
[3] S. Abbasbandy, T. Allahviranloo, P. Darabi, O. Sedaghatfar, Variational Iteration Method for Solving
N-th order Fuzzy Differential Equations, Mathematical and Computational Applications, 16 (4)
(2011).
[4] N. I. Ahmad, M. Mamat, J. Kavikumar, N. A. Hamzad, Solving Fuzzy Duffing’s Equation by the
Laplace Transform Decomposition, Applied Mathematical Sciences, 6 (59) (2012) 2935-2944.
[5] L. Assas, Variational Iteration Method for Solving Coupled-Kdv Equations, Chaos Solutions Fractals,
38 (2008) 1225-1228.
http://dx.doi.org/10.1016/j.chaos.2007.02.012
Table 4: Result and accuracy for the lower and upper bound approximate solution of
Eq. (39) at t =0.1 by 5-order VIM
r (0.1; ) (0.1, ) (0.1, ) (0.1, )
0 0.9817642960058663 1.73342 × 10 7 1.1975986372194212 2.88251 × 10 7
0.25 1.0088113021609082 1.84973 × 10 7 1.1706904673788245 2.70845 × 10 7
0.5 1.0358399373112013 1.97301 × 10 7 1.1437610043010307 2.54402 × 10 7
0.75 1.0628497128059113 2.10366 × 10 7 1.1168107333604598 2.38873 × 10 7
1 1.0628497128059113 2.24209 × 10 7 1.0628497128059113 2.24209 × 10 7
Page 12
of 1402Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
[6] M. F. Abbod, D. G. Von Keyserlingk, D. A Linkens, M. Mahfouf, Survey of Utilization of Fuzzy
Technology in Medicine and Healthcare, Fuzzy sets and system, 120 (2001) 331-349.
http://dx.doi.org/10.1016/S0165-0114(99)00148-7
[7] E. Babolian, H. Sadeghi, Sh. Javadi, Numerically Solution of Fuzzy Differential Equations by
Adomian Method, Applied Mathematics and Computation, 149 (2004) 547-557.
http://dx.doi.org/10.1016/S0096-3003(03)00160-7
[8] S. Barro, R. Marin, Fuzzy Logic in Medicine, Heidelberg: Physica-Verlag, (2002).
http://dx.doi.org/10.1007/978-3-7908-1804-8
[9] B. Berna, S. Mehmet, Numerical Solution of Duffing Equation by Using an Improved Taylor Matrix
Method, Journal of Applied Mathematics, 6 (2013).
[10] G. F. Corliss, Guarented Error Bounds for Ordinary Differential Equations in Theory and Numeric of
Ordinary and Partial Equations, Oxford University press, (1995).
[11] D. Dubois, H. Prade, Towards Fuzzy Differential Calculus, Part 3: Differentiation, Fuzzy Sets and
Systems, 8 (1982) 225-233.
http://dx.doi.org/10.1016/S0165-0114(82)80001-8
[12] D. Dubois, H. Prade, Fuzzy Numbers: An Overview, Analysis of Fuzzy Information, Mathematical
Logic, CRC Press, Boca Raton, FL, 1 (1987) 3-39.
[13] M. S. El Naschie, From Experimental Quantum Optics to Quantum Gravity via a Fuzzy Kahler
Manifold, Chaos Solution and Fractals, 25 (2005) 969-977.
http://dx.doi.org/10.1016/j.chaos.2005.02.028
[14] O. S. Fard, An Iterative Scheme for the Solution of Generalized System of Linear Fuzzy Differential
Equations, World Applied Sciences Journal, 7 (2009) 1597-11604.
[15] A. Ghorbanifar, J. Saberi-Nadjafi, Convergence of He’s Variational Iteration Method for Nonlinear
Oscillators, Nonlinear Sci. Lett. A, 1 (4) (2010) 379-384.
[16] M. Ghanbari, Numerical Solution of Fuzzy Initial Value Problems under Generalization
Differentiability by HPM, Int. J. Industrial Mathematics, 1 (1) (2009) 19-39.
[17] M. Ghanbari, Solution of the First Order Linear Fuzzy Differential Equations by Some Reliable
Methods, Journal of fuzzy set and analysis, 2012 (2012) 1-20.
http://dx.doi.org/10.5899/2012/jfsva-00126
[18] J. H. He, Some Asymptotic Methods for Strongly Nonlinear Equations, Internat. J. Modern Phys. B,
20 (2006) 1141- 1199.
http://dx.doi.org/10.1142/S0217979206033796
Page 13
of 1403Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
[19] J. H. He, Variational Iteration Method for Delay Differential Equations, Comm. Non-Linear. Sci.
Numer. Simulation, 2 (1997) 235- 236.
http://dx.doi.org/10.1016/S1007-5704(97)90008-3
[20] J. H. He, Approximate Solution of Nonlinear Differential Equations with Convolution Product Non-
Linearity’s, Comput. Methods. Appl. Mech. Engng, 167 (1998) 69-73.
http://dx.doi.org/10.1016/S0045-7825(98)00109-1
[21] J. H. He, Approximate Analytical Solution for Seepage Flow with Fractional Derivatives in Porous
Media, Comput. Methods. Appl. Mech. Engng, 167 (1998) 57-68.
http://dx.doi.org/10.1016/S0045-7825(98)00108-X
[22] J. H. He, Variational Iteration Method- A Kind of Non-Linear Analytical Technique: Some Examples,
Internat. J. Non-Linear Mech, 34 (1999) 699-708.
http://dx.doi.org/10.1016/S0020-7462(98)00048-1
[23] J. C. Jiao, Y. Yamamoto, C. Dang, Y. Ha, An after Treatment Technique for Improving the Accuracy
of Adomian’s Decomposition Method, Comput. Math. Appl, 43 (2002) 783-798.
http://dx.doi.org/10.1016/S0898-1221(01)00321-2
[24] O. Kaleva, A Note on Fuzzy Differential Equations, Nonlinear Analysis, 64 (1) (2006) 895-900.
http://dx.doi.org/10.1016/j.na.2005.01.003
[25] S. A. Khuri, A Laplace Decomposition Algorithm Applied To A Class Of Nonlinear Differential
Equations, J. Appl. Math, 1 (4) (2001) 141-155.
http://dx.doi.org/10.1155/S1110757X01000183
[26] S. Liang, J. J. David, Comparison of Homotopy Analysis Method and Technique, Applied
Mathematics and Computation, 35 (1) (2003) 73-79.
[27] M. A. Noor, K. I. Noor, S. T. Mohyud-Din, Variational Iteration Method For Solving Sixth-Order
Boundary Value Problems, Commun. Nonlinear Sci. Numer. Simulat, 14 (2009) 2571-2580.
[28] H. Nijmeijer, H. Berghuis, On Lyapunov Control of the Duffing Equation, IEEE Trans Circuits Syst.
I, 42 (1995) 473-477.
http://dx.doi.org/10.1109/81.404059
[29] W. O. Rebecca, Lower and Upper Solutions and Existence of Solutions of Fuzzy Differential
Equations, South African Journal of Pure and Applied Mathematics, 4 (2009) 29-43.
[30] S. Salahshour, Nth-order Fuzzy Differential Equations under Generalized Differentiability, Journal of
fuzzy set value analysis, 2011 (2011) 1-14.
http://dx.doi.org/10.5899/2011/jfsva-00043
Page 14
of 1404Journal of Interpolation and Approximation in Scientific Computing
http://www.ispacs.com/journals/jiasc/2014/jiasc-00053/
International Scientific Publications and Consulting Services
[31] S. Siah Mansouri, N. Ahmady, Numerical Method For solving Nth order Fuzzy Differential Equation
by using Characterization Theorem, Communication in Numerical Analysis, 2012 (2012) 1-12.
http://dx.doi.org/10.5899/2012/cna-00054
[32] N. H. Sweilam, Variational Iteration Method for Solving Cubic Nonlinear Schro Dinger Equation, J.
Comput.Appl. Math, 207 (2007) 155-163.
http://dx.doi.org/10.1016/j.cam.2006.07.023
[33] S. Seikkala, On the Fuzzy Initial Value Problem, Fuzzy Sets Syst, 24 (1987) 319-330.
http://dx.doi.org/10.1016/0165-0114(87)90030-3
[34] R. Yulita Molliq, M. S. M. Noorani, I. Hashim, Variational Iteration Method for Fractional Heat- And
Wave-Like Equations, Nonlinear Analysis: Real World Applications, 10 (3) (2009) 1854-1869.
http://dx.doi.org/10.1016/j.nonrwa.2008.02.026
[35] E. Yusufoglu, Numerical Solution of Duffing Equation by the Laplace Decomposition Algorithm,
Applied Mathematics and Computation, 177 (2) (2006) 572-580.
http://dx.doi.org/10.1016/j.amc.2005.07.072
[36] M. Zaini, Ahmad Bernard De Baets, A Predator-Prey Model with Fuzzy Initial Populations, IFSA-
EUSFLAT (2009).
[37] L. A. Zadeh, Fuzzy Sets, Information and Control, 8 (1965) 338-353.
http://dx.doi.org/10.1016/S0019-9958(65)90241-X