Applied Mathematical Sciences, Vol. 6, 2012, no.

104, 5147 - 5159

Power Series Solutions for Nonlinear Systems of Partial Differential Equations

Ameina S. Nuseir Jordan University of Science and Technology P. O. Box 3030, Irbid, 22110, Jordan anuseir@just.edu.jo Abeer Al-Hasoon Jordan University of Science and Technology P. O. Box 3030, Irbid, 22110, Jordan Abstract In this paper analytical solutions of nonlinear partial differential systems are addressed. The solutions are obtained using the technique of power series to solve linear ordinary differential equations. This method ensures the theoretical exactness of the approximate solution. Several systems are solved using this method and comparisons of the approximate solutions with the exact ones are demonstrated. Mathematics Subject Classification: 35-04, 35D99 Keywords: Nonlinear PDEs, Power Series Method, Analytical Solutions

1. Introduction
It is well known that there are several methods that can be used to find general solutions to linear PDEs. On the contrary, for non-linear PDEs it is well known that there are no generally applicable methods to solve such nonlinear equations. A glance at the literature shows that there are some known methods which have been applied to solve special cases of nonlinear PDEs. For example the split-step method is a computational method that has been used to solve specific equations like nonlinear Schrdinger equation [1, 37].

5148

A. S. Nuseir and A. Al-Hasson

Nevertheless, some techniques can be used to solve several types of nonlinear equations such as the homotopy principle which is the most powerful method to solve underdetermined equations [9]. In some cases, a PDE can be solved via perturbation analysis in which the solution is considered to be a correction to an equation with a known solution [8, 25]. Alternatively, there are numerical techniques that solve nonlinear PDEs such as the finite difference method [ 2,19,30] and the finite element methods [27,33,34,36]. Many interesting problems in science and engineering can be solved in this way using computers. A general approach to solve PDEs uses the symmetry property of differential equations, the continuous infinitesimal transformations of solutions to solutions (Lie theory) [14,26,32,35]. The continuous group theory, Lie algebras and differential geometry are used to understand the structure of linear and nonlinear partial differential equations. Then generating integrable equations to find their Lax pairs, recursion operators, Bcklund transform and finally finding exact analytic solutions to PDEs [17, 22, 23, 28, 31]. Furthermore, other direct methods were developed to find closed form solutions for nonlinear PDEs such as the Tanh method [18, 24], extended Tanh method [39], Exp-function method [15,16], rational exponential method [38], and others [40-42].

2. Power Series Method for Nonlinear Partial Differential Equations

Power series is an old technique for solving linear ordinary differential equations [7,20]. The efficiency of this standard technique in solving linear ODEs with variable coefficients is well known. An extension known as Frobenius method allows tackling differential equations with coefficients that are not analytic [3]. Recently the method has been used to solve nonlinear ODEs, [5,10-13]. Furthermore, Kurulay and Bayram [21] used power series to solve linear second order PDEs. In this work we apply the power series method to nonlinear PDEs. Analytical solutions are found by using algebraic series. Manipulation of the equations leads to very convenient recurrence relations that ensure the exactness of the solution as well as the computational efficiency of the method. The method is straightforward and can be programmed using any mathematical package. The efficiency of this method is illustrated through some examples and obtained solutions compared with exact solutions. The general algebra for solving nonlinear ODEs is explained by considering an analytical function x = x ( ) defined in {I : 0 1} . Assume its expansion in power series as

Power series solutions for nonlinear systems

5149

x( ) = a1k k and for any integer m, x( ) of power m is expressed as

x( ) m = a mk k
k =0

(1)

k =0

(2)

The following relation is an essential condition to be satisfied in order to reveal the desired recurrence relation
x m ( ) = x m 1 ( ) x( )

(3)

After replacing the series expressions in each factor of equation (3), the following recurrence relation is obtained
a mk = a ( m1) p a1( k p ) = a1 p a( m 1)( k p )
p =0 p =0 k k

(4)

To find the series expansion of a product of two functions assumes that the functions f ( ) and g ( ) are analytic at = 0 and are defined in {I : 0 1} . Let
f ( ) = ai i ; g ( ) = bi i
i =0 i =0

(5) (6)

and

h( ) = f ( ) g ( )

then the function h is also analytic at = 0 and defined in I, therefore the series expansion for h( ) is
h( ) = ci i
i =0

(7)
i

where,

ci = a s bi s = ai s bs
s =0 s =0

; i = 0,1,2,...

(8)

And the series expansion for the nth derivative x( ) , can be written as
x ( n ) ( ) = nk a ( k + n ) k
k =0

(9)

where, kj = ( j + 1)( j + 2).........( j + k ), where k and j are positive integers. For more details on solving nonlinear ODEs, see [31, 32].

5150

A. S. Nuseir and A. Al-Hasson

To generalize this method for solving nonlinear PDEs, let u( x, y ) be a function of two variables, and suppose that it is analytic in the domain G R 2 and assume that the point ( x 0 , y 0 ) in G. The function u ( x, y ) is then represented as
u ( x, y ) = aij ( x x 0 ) i ( y y 0 ) j
j =0 i =0

(10)

To find the representation series for any power of u(x, y) a condition as in (3) will be applied u m ( x, y ) = u m 1 ( x, y ) u ( x, y ) . (11) If the series expansion of u m ( x, y ) is written as
( m) u ( m ) ( x, y ) = aij ( x x0 ) i ( y y 0 ) j

(12)

j =0 i =0

then using the relation in (11), the coefficients of um expressed as

(m) aij =

s =0

a
p =0

( m 1) ps

) a ((i1 p )( j s )

(13)

A representation for any derivative of u with respect to x or y, for any order, and for any power of them, can be found by generalization the equivalent relation for ODEs. Some examples will be used to explain the method.

3. Numerical Examples
To illustrate the technique and exactness of the approximate solution, we now investigate some examples of nonlinear PDEs in detail.
Example 1. The nonlinear diffusion equation is considered m ut = (u u x ) x where m is a positive integer. Let m=2, then the equation is 2 u 2 ut = (u ) = u 2 u xx + 2uu x x x

(14)

(15)

with initial condition

Power series solutions for nonlinear systems

5151

x+h (16) 2 c where c > 0 , and h is an arbitrary constant. The exact solution of the given x+h [6]. equation is u( x, t ) = 2 ct Assume the solution u(x,t) as a power series in x and t, u( x,0) =
u ( x, t ) = aij x i t j
i =0 j =0

(17)

By differentiating both sides of (17) with respect to x, we will get the expansion series of ux and uxx
u x = (i + 1)a ( i +1), j x i t j
i =0 j =0

(18) (19)

u xx = (i + 1)(i + 2)a ( i + 2), j x i t j

i =0 j =0

2 The power series of u2, u2uxx and uu x are obtained by applying (11) and the above series
j i u 2 = ats a ( i t )( j s ) x j t j j =0 i =0 s =0 t =0

(20)

i j q p j j u 2 u xx = (i p + 1)(i p + 2)a ( i p + 2 )( j q ) a a ts p t q s ( )( ) x t (21) i =0 j =0 q =0 p =0 s =0 t =0 i j q p j j 2 = a( i p )( j q ) + + uu x ( t 1 )( p t 1 ) a a (22) t s p t q s ( 1 ) ( 1 )( ) + + x t j = 0 i = 0 q = 0 p = 0 s =0 t =0

Substitute these series into equation (11), to obtain the recurrence relation: q p 1 j i ai ( j +1) = ( i p 1 )( i p 2 ) a ats a ( p t )( q s ) + + ( i p + 2 )( j q ) + j + 1 q =0 p =0 s =0 t =0
q p 2 a ( i p )( j q ) (t + 1)( p t + 1)a ( t +1) s a ( p t +1)( q s ) q =0 p =0 s =0 t =0
j i

(23)

where

5152

A. S. Nuseir and A. Al-Hasson

ai , 0

h 2 c 1 = 2 c 0

if if

i=0 i =1 o.w

By applying the recurrence relations (23) for several values of i and j. the polynomial approximation for u(x,t) is obtained
% (x , t ) = u

h 2 c

ht 4c
3 2

3h t2 16c
5 2

5ht3 32c
7 2

x 2 c

tx 4c
3 2

3 t2 x 16c
5 2

5t 3 x 32c
7 2

(24)

Table 1 demonstrates the difference between the approximate solution and the exact one for several values of x and t when h=1 and c=10. Table 1: The difference between exact and approximate solutions for Example 1.
% t u(0,t)-u(0,t) % u(0.2,t)-u(0.2,t) % u(0.4,t)-u(0.4,t) % u(0.6,t)-u(0.6,t)

% u(0.8,t)-u(0.8,t)

% u(1,t)-u(1,t)

0 0.2 0.4 0.6 0.8 1.0

0 7.0410 1.15107 5.92107 1.91106 4.75106

9

0 8.45109 1.38107 7.11107 2.29106 5.70106

0 9.8610 1.61107 8.29107 2.67106 6.65106

9 8

0 1.1310 1.84107 9.48107 3.05106 7.60106 1.2710 2.07107 1.07106 3.44106 8.56106
8

0 1.4110 2.30107 1.18106 3.82106 9.50106

8

Example 2. Consider the nonlinear PDE

ut = u 2 u xx + u 3 + u x + u
with initial condition u( x,0) = sin x . The exact solution to (25) is u( x, t ) = e t sin( x + t ) . Assume that
u ( x, t ) = a1ij x i t j
i = 0 j =0

(25)

(26)

then
u xx = (i + 1)(i + 2)a1(i + 2 ) j x i t j
i =0 j =0

(27)

Power series solutions for nonlinear systems

5153

and
j i u 2 = a1ts a1( i t )( j s ) x j t j j =0 i =0 s =0 t =0

(28)

Therefore,
j i q p j j u 2 u xx = (i p + 1)(i p + 2)a1( i p +2 )( j q ) a1ts a1( p t )( q s ) x t (29) i =0 j =0 q =0 p =0 s =0 t =0

and,
j i s t (30) u 3 = ( a1qp a1( t q )( s p ) )a1( i t )( j s ) x j t j j = 0 i = 0 s =0 t =0 p =0 q =0 Substituting the above series into equation (25), the following recurrence equation is obtained

a1i ( j +1) =

q p 1 j i ( i p 1 )( i p 2 ) a a1ts a1( p t )( q s ) + + 1( i p + 2 )( j q ) + j + 1 q =0 p =0 s =0 t =0

j i s t ( a1qp a1( t q )( s p ) )a1( i t )( j s ) + (i + 1)a1( i +1) j + a1ij s =0 t =0 p =0 q =0 The approximate solution is computed for i = 0,1,2,3,4 and j = 0,1,2,3,4
2

(31)

% ( x,t ) = t + t 2 + u

t3 t3 x t4 x t x 2 t2 x t3 x 2 x3 t x 3 t3 x3 + x +t x + + 3 3 6 2 2 6 6 6 18 (32) 4 3 4 2 4 3 4 t x tx t x t x + + + 36 24 24 72

Table 2 below illustrates the difference between the approximate solution obtained by equation (32) and the exact solution. Table 2: The difference between exact and approximate solutions for Example 2.
t 0 0.2 0.4 0.6 0.8 1.0
u(0,t)-u(0,t) %
% u(0.2,t)-u(0.2,t) % u(0.4,t)-u(0.4,t) % u(0.6,t)-u(0.6,t) % u(0.8,t)-u(0.8,t) % u(1,t)-u(1,t)

0 2.351011 1.25108 5.01107 6.93106 5.40105

1.411012 2.461010 6.63108 1.90106 2.12105 1.40104

7.211010 5.23109 1.28107 3.25106 3.46105 2.21104

2.77108 1.34107 4.43107 4.89106 4.73105 2.94104

3.68107 1.44106 3.11106 1.03105 6.39105 3.63104

2.73106 9.22106 1.82105 3.56105 1.08104 4.59104

5154

A. S. Nuseir and A. Al-Hasson

Example 3. Consider the nonlinear system ut + vu x + u 1 = 0

vt + uv x v 1 = 0

(33)

Subject to the initial conditions u ( x,0) = e x , v( x,0) = e x . The exact solution is u( x, t ) = e x t , v ( x, t ) = e t x [4]. In order to solve the given system using the power series method, the solutions u and v are considered as
u(x,t) = ai , j x i t j
i =0 j =0 N N N N

(34) (35)

v(x,t) = bi , j x i t j
i =0 j =0

we use the representation of the solutions in equations (34) and (35) to write the power series expansion of the products vu x and uv x . Then we obtain the recursion formulas

ai , j +1 = bi , j +1 = 1 where i , j ,0 = 0

j i 1 + a ( s + 1)bi s , j t a s +1,t i , j ,0 i, j j +1 t =0 s =0

(36) (37)

j i 1 + + b ( s + 1)ai s , j t bs +1,t i , j ,0 i, j j +1 t =0 s = 0 if i = j = 0 o.w

After solving (36), and (37) for i=0,, 3 and polynomials

j=0, ,3 we obtain the

% (x , t ) = 1 t + u

t2 t3 t2 x t3 x x 2 t x 2 + x tx + + + 2 6 2 6 2 2
3

t2 x 2 t3 x 2 x3 t x3 t2 x3 t3 x + + 4 12 6 6 12 36 and,
v%(x , t ) = 1 + t +

(38)

t2 t3 t2 x t3 x x 2 t x 2 + x tx + + + 2 6 2 6 2 2
3

t2 x 2 t3 x 2 x 3 t x 3 t2 x 3 t3 x + 4 12 6 6 12 36

(39)

Power series solutions for nonlinear systems

5155

The difference between approximate solutions and the exact solutions for equation (33) are shown in tables 3 and 4.

Table 3: The difference between exact and approximate solutions of u(x,t) for Example 3.
% t u(0,t)-u(0,t) % u(0.2,t)-u(0.2,t) % u(0.4,t)-u(0.4,t) % u(0.6,t)-u(0.6,t)

% u(0.8,t)-u(0.8,t)

% u(1,t)-u(1,t)

0 0.2 0.4 0.6 0.8 1.0

0 2.48109 3.10107 5.16106 3.8105 1.8104

2.6110 8.94109 3.76107 6.31106 4.62105 2.2104

3.4210 2.76107 2.32107 7.52106 5.62105 2.63104

6.0010 4.91106 3.46106 6.12106 6.62105 3.19104

4.6210 3.78105 3.03105 1.39105 6.34105 3.75104

2.26104 1.85104 1.51104 1.10104 1.04104 3.95104

Table 4: The difference between exact and approximate solutions of v(x,t) for Example 3.
t 0 0.2 0.4 0.6 0.8 1.0
v(0,t)-v(0,t) %
% v(0.2,t)-v(0.2,t) % v(0.4,t)-v(0.4,t) % v(0.6,t)-v(0.6,t)

% v(0.8,t)-v(0.8,t)

% v(1,t)-v(1,t)

0 2.60109 3.42107 6.00106 4.61105 2.26104

2.4810 8.941010 2.76107 4.91106 3.78105 1.85104

3.1010 3.76107 2.32107 3.46106 3.03105 1.51104

5.1610 6.31106 7.52106 6.12106 1.38105 1.10104

3.7810 4.62105 5.62105 6.62105 6.34105 1.05104

1.76104 2.15104 2.63104 3.19104 3.75104 3.95104

Example 4. The coupled Burgers equation ut u xx 2uu x + (uv ) x = 0

vt v xx 2vv x + (uv ) x = 0

(40)

with initial conditions u( x,0) = sin x , v( x,0) = sin x . The exact solution of the above system is u(x,t )= v(x,t )= sinx e-t [29]. In order to solve the given system using the PSM, the solutions u and v is considered as
u(x,t) = a i , j x i t j
i =0 j =0

(41) (42)

v(x,t) = bi , j x i t j
i =0 j =0

5156

A. S. Nuseir and A. Al-Hasson

After substitution the appropriate series in the equation, we get the following recurrence relations j i 1 ai , j +1 = ( i 1 )( i 2 ) a 2( s + 1)ai s , j t a s +1,t + + + i + 2, j j +1 t =0 s =0 (43) j j i i + ( s + 1)ai s , j t bs +1,t ( s + 1)bi s , j t a s +1,t t =0 s = 0 t =0 s =0 j i 1 bi , j +1 = (i + 1)(i + 2)bi + 2, j + 2( s + 1)bi s , j t bs +1,t j +1 t =0 s =0 (44) j j i i + ( s + 1)ai s , j t bs +1,t ( s + 1)bi s , j t a s +1,t t =0 s = 0 t =0 s =0 Then the desired coefficients are obtained by repeated application of the recurrence relations. Using mathematica program to solve the preceding recurrence system, for i=0, ,3 and j =0, ,3. We obtain the following polynomials for u(x,t) and v(x,t) t 3 x x 3 tx 3 2t 3 x 3 + + 3 6 6 9 3 3 3 2t 3 x 3 t x x tx v% (x , t ) = x tx + + 3 6 6 9 % (x , t ) = x tx u

(45)

Since the exact solutions are the same, then the approximate solutions u(x,t) and v(x,t) are also the same. Tables 5 and 6 show the convergence of the approximate solutions to the exact ones for different values of x and t.

Table 5: The difference between exact and approximate solutions of u(x,t) for Example 4.
% t u(0,t)-u(0,t) % u(0.2,t)-u(0.2,t) % u(0.4,t)-u(0.4,t) % u(0.6,t)-u(0.6,t)

% u(0.8,t)-u(0.8,t)

% u(1,t)-u(1,t)

0 0.2 0.4 0.6 0.8 1.0

0 0 0 0 0 0

1.4110 1.351011 3.09109 7.76108 7.59107 4.43106

12

7.2110 6.151010 6.54109 1.52107 1.49106 8.68106

10

2.7910 2.27108 2.73108 2.36107 2.17106 1.26105

3.6810 3.01107 2.58107 4.82107 2.90106 1.61105

2.73106 2.24106 1.84106 1.83106 4.44106 1.98105

Power series solutions for nonlinear systems

5157

Table 6: The difference between exact and approximate solutions of v(x,t) for Example 4.
% t v(0,t)-v(0,t) % v(0.2,t)-v(0.2,t) % v(0.4,t)-v(0.4,t) % v(0.6,t)-v(0.6,t)

% v(0.8,t)-v(0.8,t)

% v(1,t)-v(1,t)

0 0.2 0.4 0.6 0.8 1.0

0 0 0 0 0 0

1.4110 1.351011 3.09109 7.76108 7.59107 4.43106

12

7.2110 6.151010 6.54109 1.52107 1.49106 8.68106

10

2.7910 2.27108 2.73108 2.36107 2.17106 1.26105

3.6810 3.01107 2.58107 4.82107 2.90106 1.61105

2.73106 2.24106 1.84106 1.83106 4.44106 1.98105

Conclusion
The method has been successfully applied directly to some examples of nonlinear PDEs without using linearization, perturbation, or restrictive assumptions. It provides the solution in terms of convergent series with easily computable components and the results have shown remarkable performance. The efficiency of this method has been demonstrated by solving some nonlinear PDEs and some systems of nonlinear PDEs. A comparison of this method with the exact solutions were performed and presented.

References
[1] G. P. Agrawal, Nonlinear Fiber Optics, Academic Press, New York (2001). [2] W. F. Ames, Numerical Methods for Partial Differential Equations, Academic Press, New York (1977). [3] G. Arfken, H. Weber, Mathematical Methods for Physicists, Academic Press, New York (1985). [4] A. S. Bataineh, M.S. Noorani, and I. Hashim, Approximate analytical solutions of systems of PDEs by homotopy analysis method. Comput. Math. Appl., 55: 2913-2923 (2008). [5] J. Biazar, M. Ilie, A. Khoshkenar, A new approach to the solution of the prey and predator problem and comparison of the results with the Adomian method, Appl. Math. Comput. 171, 486491 (2005). [6] C. Chun, H. Jafari, Y. Kim, Numerical method for the wave and the nonlinear diffusion equations with the homotopy perturbation method. Comput. Math. Appl., 57: 1226-1231 (2009). [7] E.A. Coddington; An Introduction to Ordinary Differential Equations, Dover Publications, New York (1989). [8] W. H. Cropper, Great Physicists: The Life and Times of Leading Physicists from Galileo to Hawking, Oxford University Press (2004) [9] Y. Eliashberg, N. M. Mishachev, Introduction to the h-principle, American Mathematical Soc. (2002).

5158

A. S. Nuseir and A. Al-Hasson

[10] V. Fairen, V. Lopez, and L. Conde; Power series approximation to solutions of nonlinear systems of differential equations, American Journal of Physics; 56, 57-61 (1988). [11] C.P. Filipich, M.B. Rosales; A Straightforward approach to solve ordinary nonlinear differential systems, Mecanica Computacional; Vol.XXI; 1549-1568 (2002). [12] C. P. Filipich, M.B. Rosales, and F. Buezas, Some nonlinear mechanical problems solved with analytical solutions, J. Latin American Appl. Research; 34, 101-109 (2004). [13] N. Guzel, M. Bayram, Power series solution of non-linear first order differential equation systems, Trakya Univ J. Sci, 6(1): 107-111 (2005). [14] T. Hawkins, Emergence of the Theory of Lie Groups: an Essay in the History of Mathematics, Springer (2000). [15] J. H. He, X. H. Wu, Exp-function method for nonlinear wave equations, Chaos, Solitons Fractals, 30, 700708 (2006). [16] J.H. He, L.N. Zhang, Generalized solitary solution and compaction-like solution of the JaulentMiodek equations using the Exp-function method, Phys. Lett. A, 372, 10441047 (2008). [17] R. Hermann, The Geometry of Non-Linear Differential Equations, Backlund Transformations, and Solitons, Part A (Interdisciplinary Mathematics Series No. 12): Math Science Press (1976). [18] W. Hereman and W. Malfliet, The tanh method: a tool to solve nonlinear partial differential equations with symbolic software, Proceedings 9th World Multi-Conference on Systemics, Cybernetics and Informatics, Orlando, FL, 165 168 (2005). [19] F. B. Hildebrand, Finite-Difference Equations and Simulations, PrenticeHall, New Jersey (1968). [20] E. Kreyszig; Advanced Engineering Mathematics, Wiley & Sons, New York (1999). [21] M. Kurulay and M. Bayram; A novel power series method for solving second order partial differential equations, European Journal Of Pure And Applied Mathematics Vol. 2, No. 2, 268-277 (2009). [22] P. Lax, Integrals of nonlinear equations of evolution and solitary waves, Comm. Pure Applied Math. 21, 467490 (1968). [23] P. Lax and R.S. Phillips, Scattering Theory for Automorphic Functions, Princeton University Press (1976). [24] W. Malfliet, Solitary wave solutions of nonlinear wave equations, Amer. J. Phys. 60 650654 (1992). [25] A. H. Nayfeh, Perturbation Methods, Wiley & Sons, New York (1973) [26] P.J. Olver, Equivalence, Invariants and Symmetry, Cambridge Press (1995). [27] G. Pelosi, The finite-element method, Part I: R. L. Courant: Historical Corner, Antennas and Propagation Magazine, IEEE, 180-182 (2007). [28] A. D. Polyanin and V. F. Zaitsev, Handbook of Nonlinear Partial Differential Equations, Chapman & Hall, CRC Press (2004).

Power series solutions for nonlinear systems

5159

[29] A.S. Ravi Kanth, K. Aruna, Differential transform method for solving linear and non-linear systems of partial differential equations. Phys. Lett. A., 372, 68966898 (2008). [30] R. D. Richtmyer and K. W. Morton, Difference Methods for Initial Value Problems, 2nd ed., Wiley, New York (1967). [31] C. Rogers, W.F. Shadwick, Bcklund transformations and their applications, Academic Press, New York (1982). [32] D. H. Sattinger, O.L. Weaver, Lie Groups And Algebras With Applications To Physics, Geometry, And Mechanics, Springer-Verlag (1986). [33] P. Solin, K. Segeth, I. Dolezel, Higher-Order Finite Element Methods, Chapman & Hall, CRC Press (2003). [34] E. Stein, O.C. Zienkiewicz, A pioneer in the development of the finite element method in engineering science, Steel Construction, 2 (4), 264-272 (2009). [35] H. Stephani, Differential Equations: Their Solution Using Symmetries, Edited by M. MacCallum, Cambridge University Press (1989). [36] G. Strang, G. Fix, An Analysis of the Finite Element Method, Prentice Hall (1973). [37] T. R. Taha and M. J. Ablowitz. Analytical and numerical aspects of certain nonlinear evolution equations. II. Numerical, nonlinear Schrdinger equation, J. Comput. Phys. 55 (2), 203230 (1984). [38] A M Wazwaz, The tanhcoth method for new compactons and solitons solutions for the K(n,n) and the K(n + 1,n + 1) equations, Appl. Math. and Comp., 188, 19301940 (2007). [39] Wazwaz A. M., The extended tanh method for new solitons solutions for many forms of the fifth-order KdV equations, Appl. Math. and Comp., 184, 10021014 (2007). [40] Wazwaz A. M., Partial differential equations: methods and applications. The Netherlands: Balkema Publishers; 2002. [41] Wazwaz A. M., A computational approach to soliton solutions of the Kadomtsev-Petviashvili equation, Appl. Math. and Comp., 123, 205-217(2001) [42] Wazwaz A. M., The tanh and the sinecosine methods for a reliable treatment of the modified equal width equation and its variants, Commun. Nonlinear Sci. Numer. Simulat., 11(2), 148-160 (2006).

Received: April, 2012