Computers and Mathematics with Applications 69 (2015) 804–816

Contents lists available at ScienceDirect

Computers and Mathematics with Applications

journal homepage: www.elsevier.com/locate/camwa

′
The compound ( GG )-expansion method and double
non-traveling wave solutions of (2 + 1)-dimensional
nonlinear partial differential equations✩
Shimin Guo a,∗ , Liquan Mei a , Yubin Zhou b
a
School of Mathematics and Statistics, Xi’an Jiaotong University, Xi’an, 710049, China
b
School of Mathematics and Statistics, Lanzhou University, Lanzhou, 730000, China

article info abstract

Article history: To seek the exact double non-traveling wave solutions of nonlinear partial differential
Received 22 August 2014 ′
equations, the compound ( GG )-expansion method is firstly proposed in this paper. With
Received in revised form 3 February 2015
the aid of symbolic computation, this new method is applied to construct double non-
Accepted 14 February 2015
Available online 13 March 2015 traveling wave solutions of (2 + 1)-dimensional Painlevé integrable Burgers equation and
(2 + 1)-dimensional breaking soliton equation. As a result, abundant double non-traveling
Keywords: wave solutions including double hyperbolic function solutions, double trigonometric func-
′
Compound ( GG )-expansion method
tion solutions, double rational solutions, and a series of complexiton solutions of these two
Double non-traveling wave solution equations are obtained via the proposed method. These exact solutions contain arbitrary
(2 + 1)-dimensional Painlevé integrable functions, which may be helpful to explain some complex phenomena. When the parame-
Burgers equation ters are taken as special values, the double solitary-like wave solutions can be derived from
(2 + 1)-dimensional breaking soliton double hyperbolic function solutions. Furthermore, the time evolutions of double solitary-
equation like wave solutions are discussed in detail.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

In the present paper, we will seek double non-traveling wave solutions of the following two systems of nonlinear partial
differential equations (NPDEs): one is the (2 + 1)-dimensional Painlevé integrable Burgers (PIB) equation [1,2] in the form

ut = uuy + αv ux + β uyy + αβ uxx ,


(1)
ux = v y ,
where α and β are nonzero constants. The above equation is quite famous in wave theory, which has a wide application in
gas dynamics and in plasma physics. The other one is the (2 + 1)-dimensional breaking soliton (BS) equation [3,4] in the form

ut + β uxxy + 4β(uv)x = 0,

(2)
uy = v x ,
where β is a nonzero constant. Eq. (2) describes the (2 + 1)-dimensional interaction of a Riemann wave propagating along
the y-axis with a long wave along the x-axis.

✩ The project is supported by National Natural Science Foundation of China (11371289, 11371288), China Postdoctoral Science Foundation
(2014M560756), and Fundamental Research Funds for the Central Universities.
∗ Corresponding author. Tel.: +86 29 82669051.
E-mail addresses: [email protected] (S. Guo), [email protected] (L. Mei), [email protected] (Y. Zhou).

http://dx.doi.org/10.1016/j.camwa.2015.02.016
0898-1221/© 2015 Elsevier Ltd. All rights reserved.
 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816 805

Both Eqs. (1) and (2) are very important in the field of mathematical physics. For a better understanding of the nonlinear
mechanics, it is a significant task to find exact solutions of these two equations. Up to now, a vast variety of methods
have been established and developed to construct exact solutions of NPDEs, such as the inverse scattering method [5],
truncated Painlevé expansion [6], Bäcklund transformation [7], Darboux transformation [8], Lie symmetries method [9],
and homogeneous balance method [10].
In recent years, direct methods [11–21] which can be used to construct exact solutions of NPDEs have become more and
more attractive partly due to the availability of symbolic computation packages like Maple and Mathematica, which enables
us to perform the tedious and complex computation on computer. The exact solutions of NPDEs can be obtained by using
′
a set of solutions of integrable ordinary differential equations [22–24]. Among these direct methods, the ( GG )-expansion
method which was first proposed in Ref. [25] is effective to construct exact solutions of NPDEs. By means of this method,
Wang et al. successfully obtained the traveling wave solutions of four kinds of NPDEs. Later, this method was improved in
Refs. [26–28] and applied to a series of NPDEs successfully.
However, we find that these methods can only get single solitary wave solutions and single traveling wave solutions
′
of NPDEs. A natural question is that: Could the ( GG )-expansion method be developed to construct multiple non-traveling
solutions including multiple solitary-like wave solutions and complexiton solutions [29] of NPDEs?
′
To answer the above question, we propose a new method called compound ( GG )-expansion method to construct double
non-traveling wave solutions of NPDEs in this paper. For illustration, we apply the new method to the (2+1)-dimensional PIB
equation (1) and the (2 + 1)-dimensional BS equation (2). As a result, a rich variety of exact solutions which include double
hyperbolic function solutions, double solitary-like wave solutions, double triangular function solutions, double rational
solutions and complexiton solutions of the above two equations are obtained via the proposed method. In Ref. [29], Ma
defined and found the complexiton solutions of KdV equation by the bilinear form. In the present work, we obtain the
complexiton solutions of NPDEs in a more direct way.
′
The rest of this paper is organized as follows. In Section 2, a general theory of the compound ( GG )-expansion method
to construct double non-traveling wave solutions of NPDEs is described. In Section 3, we illustrate the applications of this
method to the (2 + 1)-dimensional PIB equation and (2 + 1)-dimensional BS equation. Finally, conclusions are given in
Section 4.
′
2. Description of the compound ( GG )-expansion method

′
In this section, we describe the main steps of the compound ( GG )-expansion method for double non-traveling wave
solutions of NPDEs.
Step 1: Consider a given NPDE with independent variables X = (x1 , x2 , . . . , t ) and dependent variable u:
P (u, ut , ux1 , ux2 , . . . , ux1 t , ux2 t , . . . , utt , ux1 x1 , ux2 x2 , . . .) = 0, (3)
where u = u(x1 , x2 , . . . , t ) is an unknown function, P is a polynomial in u(x1 , x2 , . . . , t ) and its various partial derivatives,
in which the highest order derivatives and nonlinear terms are involved.
Step 2: We suppose that Eq. (3) has the following solution

G′1 (ξ ) G′2 (η)

n 
  i  j
u(x1 , x2 , . . . , t ) = a0 + aij , (4)
k=1 i+j=k
G1 (ξ ) G2 (η)

where ξ , η, a0 , aij (i, j = 0, 1, 2, . . .) are functions of X = (x1 , x2 , . . . , t ) to be determined later, G1 = G1 (ξ ) and G2 =

G2 (η) satisfy the following second order linear ordinary differential equations (LODEs)

G′′1 + λ1 G′1 + µ1 G1 = 0, G′′2 + λ2 G′2 + µ2 G2 = 0, (5)

where λ1 , λ2 , µ1 , µ2 are constants, and (λ1 − λ2 ) + (µ1 − µ2 ) ̸= 0. 2 2

Step 3: Determine the positive integer n by balancing the highest order derivatives and nonlinear terms in Eq. (3) (see
Refs. [10,11] for details).
G′ G′
Step 4: Substituting ansätz (4) along with Eq. (5) into Eq. (3) and collecting all the terms with the same order of ( G1 )i ( G2 )j
1 2
G′ G′
(i, j = 0, 1, 2, . . .), we convert the left-hand side of Eq. (3) into a polynomial in ( G11 )( G22 ). Vanishing all the coefficients of
this polynomial, we obtain a set of over-determined partial differential equations for ξ , η, a0 , aij (i, j = 0, 1, 2, . . .).
Step 5: Assuming that the functions ξ , η, a0 , aij (i, j = 0, 1, 2, . . .) can be obtained by solving the over-determined
partial differential equations in Step 4, and substituting these functions and the known general solutions of Eq. (5) into
ansätz (4), we can obtain the double non-traveling wave solutions of Eq. (3) immediately.
′
3. Applications of the compound ( GG )-expansion method

In this section, we will construct double non-traveling wave solutions of (2 + 1)-dimensional PIB equation (1) and (2 + 1)-
′
dimensional BS equation (2) by using the compound ( GG )-expansion method described in Section 2.
806 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816

3.1. The double non-traveling wave solutions of (2 + 1)-dimensional Painlevé integrable Burgers equation

By balancing the highest order derivative terms and nonlinear terms in Eq. (1), we suppose that Eq. (1) own the solutions
in the form
 ′  ′
G1 G2


 u = a 0 + a 10 + a 01 ,
G1 G2

 ′  ′ (6)
G1 G2
v = b0 + b10 ,

 + b01
G1 G2

where G1 = G1 (ξ ), G2 = G2 (η) satisfy Eq. (5), and ξ , η, a0 , a10 , a01 , b0 , b10 , b01 are functions of (x, y, t ) to be determined
later.
G′ G′
Substituting (6) along with (5) into Eq. (1), and vanishing all the coefficients of ( G1 )i ( G2 )j , we obtain a set of over-
1 2
determined partial differential equations for ξ , η, a0 , a10 , a01 , b0 , b10 , b01 . Solving the over-determined partial differential
equations by Maple software, we can obtain the following results

ξ = C1 x + Py + α C1 ϕ(t )dt + (−λ1 β P 2 + C2 P − λ1 αβ C12 )t + C3 , (7)

PQ ϕ(t ) λ2 β P 2 Q 2
  
PQx
η=− + Qy − dt + − − λ2 β Q + QC2 t + C4 ,
2
(8)
α C1 C1 α C12
a0 = C2 , a10 = 2β P , a01 = 2β Q , (9)
β PQ
b0 = ϕ(t ), b10 = 2β C1 , b01 = −2 , (10)
α C1
where ϕ(t ) is an arbitrary function of t, P , Q , C1 , C2 , C3 , and C4 are arbitrary constants. Using (7)–(10), we can rewrite
expression (6) as
 ′  ′
G1 G2


 u = C 2 + 2β P + 2β Q ,
G1 G2

(11)
β PQ G2
 ′  ′
G1
v = ϕ(t ) + 2β C1 .

 −2
G1 α C1 G2
Substituting the general solutions of Eq. (5) into Eq. (11), we can obtain the double non-traveling wave solutions of (2 + 1)-
dimensional PIB equation
 (1) as follows.
λ21 − 4µ1 > 0
Case 1: When , we obtain the double hyperbolic function solution
λ22 − 4µ2 > 0

      


  A1 sinh 2 λ1 − 4µ1 ξ + A2 cosh 2 λ1 − 4µ1 ξ
1 2 1 2

u1 (x, y, t ) = β P λ21 − 4µ1




      

A cosh 1
λ 2
− 4 µ ξ + A sinh 1
λ 2
− 4µ ξ



 1 2 1 1 2 2 1 1



      

B1 sinh 2 λ2 − 4µ2 η + B2 cosh 2 λ2 − 4µ2 η
1 2 1 2



 
β λ 2
µ  + C2 − λ1 β P − λ2 β Q ,




 + Q 2 − 4 2     
1
λ 2
µ η 1
λ2
µ η



 B 1 cosh 2 2 − 4 2 + B 2 sinh 2 2 − 4 2

      (12)
A1 sinh 2 λ1 − 4µ1 ξ + A2 cosh 2 λ1 − 4µ1 ξ
1 2 1 2



 

v ( , , ) = β λ 2
− µ



 1 x y t C 1 1 4 1      
1
λ 2
µ ξ 1
λ 2
µ ξ




 A 1 cosh 2 1 − 4 1 + A 2 sinh 2 1 − 4 1



      
B1 sinh 12 λ22 − 4µ2 η + B2 cosh 12 λ22 − 4µ2 η


β PQ β PQ

 

λ2 − 4µ2
2
 + ϕ(t ) − λ1 β C1 + λ2 ,

 − αC

α C1
     
1
B1 cosh 21 λ22 − 4µ2 η + B2 sinh 12 λ22 − 4µ2 η




where A1 , A2 , B1 , B2 , P , Q , C1 , C2 are arbitrary constants, ξ and η satisfy (7) and (8), respectively.
 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816 807

In particular, setting A1 ̸= 0, B2 ̸= 0, A2 = B1 = 0, we can rewrite the solution (12) as

  
1
 
 u 11 ( x , y , t ) = β P λ2
1 − 4 µ 1 tanh λ 2
1 − 4µ 1 ξ
2



   
1
 
 + β Q λ2 − 4µ coth

λ2 − 4µ2 η + C2 − λ1 β P − λ2 β Q ,
2
2
 2
2

(13)
β PQ
     
1 1
 
v ( , , ) β λ µ λ µ ξ λ µ λ µ η
 2 2 2 2

 11 x y t = C 1 − 4 1 tanh − 4 1 − − 4 2 coth − 4 2
1
2 1
α C1 2
2 2



β PQ


 + ϕ(t ) − λ β C + λ

.
1 1 2
α C1
which is the double solitary-like wave solution of Eq. (1).
Setting again A1 > 0, B1 > 0, A21 > A22 , B21 > B22 in the solution (12), we obtain the following double solitary-like wave
solution of (2 + 1)-dimensional PIB equation (1)
     
1 1
  

 u 12 ( x , y , t ) = β P λ2
1 − 4 µ 1 tanh λ 2
1 − 4 µ 1 ξ + ξ 0 + β Q λ 2
2 − 4 µ 2 tanh λ2
2 − 4 µ 2 η + η 0

 2 2
λ β λ β ,



 + C 2 − 1 P − 2 Q
β PQ (14)
     
1 1
 
v
 12 (x , y, t ) = β C 1 λ2
− 4µ tanh λ 2
− 4µ 1 ξ + ξ0 − λ 2
− 4µ 2 tanh λ 2
− 4µ 2 η + η 0
α C1
 1 1 2 2
 2 2
β

PQ


 + ϕ(t ) − λ β C + λ

,
1 1 2
α C1
   
where ξ0 = tanh−1 A2 , η0 = tanh−1 B2 .
A B
1 1

Owing to the arbitrary function ϕ(t ) in the double hyperbolic function solution (12), it is convenient for us to excite
abundant nonlinear structures of (2 + 1)-dimensional PIB equation (1). In Fig. 1, we present the double kink solitary-like
wave profiles of quantities u12 and v12 which are expressed by Eq. (14). Clearly, this kink solution is composed of two single
kink solutions. It is interesting to point out that these two single kink solutions have the same shock front, which has never
been reported in the previous literature. Fig. 2 displays the time evolutions of u12 and v12 for fixed y. From plot (a), we can see
that the amplitude of u12 decreases with an increase of time. Plot (b) shows the kink periodic wave solution of v12 , because
we set the arbitrary functions ϕ(t ) as the periodic function, i.e., ϕ(t ) = sin2 (t ).
It is easy to see that if A1 , A2 , B1 , B2 , λ1 , λ2 , µ1 , µ2 , and the arbitrary functions ϕ(t ) are taken as other special values
in a proper way, more double solitary-like wave solutions of Eq. (1) can be obtained from the double hyperbolic function
solution (12). Here, we omit them for simplicity.

λ21 − 4µ1 < 0
Case 2: When , we obtain double trigonometric function solution
λ22 − 4µ2 < 0

      

  A1 cos 2 4µ1 − λ1 ξ − A2 sin 2 4µ1 − λ1 ξ
1 2 1 2

( , , ) β µ λ

2
u x y t = P 4 −



 2 1 1
     
1
µ λ2
ξ 1
µ λ 2
ξ




 A 1 sin 2
4 1 − 1 + A2 cos 2
4 1 − 1

      


B1 cos 12 4µ2 − λ22 η − B2 sin 21 4µ2 − λ22 η



 
+ β Q 4µ2 − λ2  + C2 − λ 1 β P − λ 2 β Q ,

 2

     

B1 sin 12 4µ2 − λ22 η + B2 cos 21 4µ2 − λ22 η




      (15)
1
µ λ 2
ξ 1
µ λ 2
ξ



  A 1 cos 2
4 1 − 1 − A 2 sin 2
4 1 − 1
v ( , , ) β µ λ
 2
x y t = C 4 −

2 1 1

1

      
A1 sin 12 4µ1 − λ21 ξ + A2 cos 21 4µ1 − λ21 ξ







      

1
µ λ 2
η 1
µ λ 2
η

 B 1 cos 4 2 − − B 2 sin 4 2 −
β PQ 2 2 2 2
β PQ

 
4µ2 − λ22  + ϕ(t ) − λ1 β C1 + λ2 ,

 −


α α C1
 
C
 
1
B1 sin( 2 4µ2 − λ2 η) + B2 cos 2 4µ2 − λ2 η
1 2 1 2



where A1 , A2 , B1 , B2 , P , Q , C1 , C2 are arbitrary constants, ξ and η satisfy (7) and (8), respectively.
808 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816

Fig. 1. The double solitary-like wave solution for the fields u12 (a) and v12 (b) expressed by Eq. (14) with P = Q = C1 = C2 = C3 = C4 = α = β = 1,
λ1 = 1, λ2 = 2, µ1 = 0.1, µ2 = 0.5, A1 = 10, A2 = 2, B1 = 10, B2 = 1, ϕ(t ) = sin2 (t ) at the fixed time t = 5.

Fig. 2. The evolutional plots of double solitary-like structures u12 (a) and v12 (b) expressed by Eq. (14) with the same parameters as in Fig. 1.


λ21 − 4µ1 > 0
Case 3: When , we get the complexiton solution combining hyperbolic and trigonometric function
λ22 − 4µ2 < 0

      


  A1 sinh 2 λ1 − 4µ1 ξ + A2 cosh 2 λ1 − 4µ1 ξ
1 2 1 2

u3 (x, y, t ) = β P λ21 − 4µ1




      

1
λ 2
µ ξ 1
λ 2
µ ξ



 A 1 cosh 2 1 − 4 1 + A 2 sinh 2 1 − 4 1

      


B1 cos 21 4µ2 − λ22 η − B2 sin 12 4µ2 − λ22 η



 
 + β Q 4µ2 − λ2 2
 + C2 − λ1 β P − λ2 β Q ,


     

1
µ λ 2
η 1
µ λ 2
η



 B 1 sin 2
4 2 − 2 + B 2 cos 2
4 2 − 2

      
(16)

  A1 sinh 1
2
λ 2
1 − 4 µ 1 ξ + A 2 cosh 1
2
λ 2
1 − 4 µ 1 ξ
v3 (x, y, t ) = β C1 λ21 − 4µ1



      

1
λ 2
µ ξ 1
λ 2
µ ξ



 A 1 cosh 2 1 − 4 1 + A 2 sinh 2 1 − 4 1







      
1
µ λ 2
η 1
µ λ 2
η

 B1 cos 4 2 − − B 2 sin 4 2 −
 − β PQ 4µ2 − λ2 β PQ
 2 2

  2 2
 + ϕ(t ) − λ1 β C1 + λ2 ,

α 2
α C1
     
C

1
B1 sin 12 4µ2 − λ22 η + B2 cos 21 4µ2 − λ22 η




where A1 , A2 , B1 , B2 , P , Q , C1 , C2 are arbitrary constants, ξ and η satisfy (7) and (8), respectively.
 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816 809

λ21 − 4µ1 < 0
Case 4: When , we obtain the complexiton solution combining hyperbolic and trigonometric function
λ22 − 4µ2 > 0

      

  A1 cos 2 4µ1 − λ1 ξ − A2 sin 2 4µ1 − λ1 ξ
1 2 1 2

( , , ) β µ λ

2
u x y t = P 4 −



 4 1 1
     
1
µ λ2
ξ 1
µ λ 2
ξ




 A 1 sin 2
4 1 − 1 + A2 cos 2
4 1 − 1

      


B1 sinh 12 λ22 − 4µ2 η + B2 cosh 21 λ22 − 4µ2 η



 
 + β Q λ2 − 4µ2  + C2 − λ 1 β P − λ 2 β Q ,

 2

     
B1 cosh 12 λ22 − 4µ2 η + B2 sinh 12 λ22 − 4µ2 η




      (17)
1
µ λ 2
ξ 1
µ λ 2
ξ



  A 1 cos 2
4 1 − 1 − A 2 sin 2
4 1 − 1
v ( , , ) β µ λ
 2
x y t = C 4 −

4 1 1

1

      
A1 sin 12 4µ1 − λ21 ξ + A2 cos 21 4µ1 − λ21 ξ







      

1
λ2
µ η 1
λ 2
µ η

 B 1 sinh − 4 2 + B 2 cosh − 4 2
β PQ 2 2 2 2
β PQ

 
λ22 − 4µ2  + ϕ(t ) − λ1 β C1 + λ2 ,

 −


α α C1
    
 C 1
B1 cosh 2 λ2 − 4µ2 η + B2 sinh 2 λ2 − 4µ2 η
1 2 1 2



where A1 , A2 , B1 , B2 , P , Q , C1 , C2 are arbitrary constants, ξ and η satisfy (7) and (8), respectively.

λ21 − 4µ1 > 0
Case 5: When , we obtain the complexiton solution combining hyperbolic and rational function
λ22 − 4µ2 = 0

      
 A1 sinh 1
λ2
− 4 µ 1 ξ + A2 cosh 1
λ2
− 4 µ 1 ξ
2 1 2 1
2β QB2

 
u5 (x, y, t ) = β P λ21 − 4µ1

 +
1 + B2 η
   
B

  
A1 cosh 12 λ21 − 4µ1 ξ + A2 sinh 12 λ21 − 4µ1 ξ






 + C2 − λ1 β P − λ2 β Q ,



(18)
     
A 1 sinh 1
λ2
− 4µ 1 ξ + A 2 cosh 1
λ 2
− 4µ 1 ξ
2β PQB2

  2 1 2 1
v5 (x, y, t ) = β C1 λ21 − 4µ1

−


α 1 (B1 + B2 η)
    
 C
A1 cosh 21 λ21 − 4µ1 ξ + A2 sinh 21 λ21 − 4µ1 ξ






 + ϕ(t ) − λ1 β C1 + λ2 β PQ ,




α C1
where A1 , A2 , B1 , B2 , P , Q , C1 , C2 are arbitrary constants, ξ and η satisfy (7) and (8), respectively.

λ21 − 4µ1 < 0
Case 6: When , we get the complexiton solution combining trigonometric and rational function
λ22 − 4µ2 = 0

      
 A1 cos 1
4µ 1 − λ 2
ξ − A 2 sin 1
4µ 1 − λ 2
ξ
2 1 2 1
2β QB2

 
u6 (x, y, t ) = β P 4µ1 − λ21

 +
1 + B2 η
   
B

  
A1 sin 12 4µ1 − λ21 ξ + A2 cos 12 4µ1 − λ21 ξ






 + C2 − λ1 β P − λ2 β Q ,



(19)
     
A1 cos 1
4µ 1 − λ 2
ξ − A 2 sin 1
4µ 1 − λ 2
ξ
2β PQB2

  2 1 2 1
v6 (x, y, t ) = β C1 4µ1 − λ21

−


α 1 (B1 + B2 η)
    
 C
A1 sin 12 4µ1 − λ21 ξ + A2 cos 21 4µ1 − λ21 ξ






 + ϕ(t ) − λ1 β C1 + λ2 β PQ ,




α C1
where A1 , A2 , B1 , B2 , P , Q , C1 , C2 are arbitrary constants, ξ and η satisfy (7) and (8), respectively.
810 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816

λ21 − 4µ1 = 0
Case 7: When , we obtain the complexiton solution combining hyperbolic and rational function
λ22 − 4µ2 > 0

      
B1 sinh 1
λ 2
− 4 µ 2 η + B 2 cosh 1
λ2
− 4 µ 2 η
2β PA2 2 2 2 2

 
( , , ) β λ µ
 2
u x y t = + Q − 4

7 2

A1 + A2 ξ 2

      
B1 cosh 2 λ2 − 4µ2 η + B2 sinh 2 λ2 − 4µ2 η
1 2 1 2






 + C2 − λ 1 β P − λ 2 β Q ,




      
B1 sinh 1
λ 2
− 4 µ 2 η + B 2 cosh 1
λ 2
− 4 µ 2 η (20)
2β C1 A2 β PQ 2 2 2 2
 
v7 (x, y, t ) = λ22 − 4µ2

 −
A1 + A2 ξ α C1
      
1
λ µ η 1
λ µ η
 2 2
B cosh − 4 + B sinh − 4

1 2 2 2
 2 2

 2 2





β PQ


+ ϕ(t ) − λ1 β C1 + λ2 ,


α C1
where A1 , A2 , B1 
, B2 , P , Q , C1 , C2 are arbitrary constants, ξ and η satisfy (7) and (8), respectively.
λ21 − 4µ1 = 0
Case 8: When , we obtain the complexiton solution combining trigonometric and rational function
λ22 − 4µ2 < 0

      
 B 1 cos 1
4µ 2 − λ 2
η − B 2 sin 1
4 µ 2 − λ 2
η
2β PA2 2 2 2 2

 
u8 ( x , y , t ) = + β Q 4µ2 − λ22


A1 + A2 ξ

      

B1 sin 2 4µ2 − λ2 η + B2 cos 2 4µ2 − λ2 η
1 2 1 2






 + C2 − λ1 β P − λ2 β Q ,



(21)
     
B1 cos 1
4µ 2 − λ 2
η − B2 sin 1
4 µ 2 − λ 2
η
2β C1 A2 β PQ

  2 2 2 2
v8 (x, y, t ) = 4µ2 − λ22

−


A1 + A2 ξ α C1

      
B1 sin 2 4µ2 − λ2 η + B2 cos 2 4µ2 − λ2 η
1 2 1 2






 + ϕ(t ) − λ1 β C1 + λ2 β PQ ,




α C1
where A1 , A2 , B1 
, B2 , P , Q , C1 , C2 are arbitrary constants, ξ and η satisfy (7) and (8), respectively.
λ21 − 4µ1 = 0
Case 9: When , we get the double rational solution
λ22 − 4µ2 = 0

2β PA2 2β QB2

u9 (x, y, t ) = A + A ξ + B + B η + C2 − λ1 β P − λ2 β Q ,


1 2 1 2
(22)
2β C1 A2 2β PQB2 β PQ
v9 (x, y, t ) = + ϕ(t ) − λ1 β C1 + λ2 .

 −
A1 + A2 ξ α C1 (B1 + B2 η) α C1
where A1 , A2 , B1 , B2 , P , Q , C1 , C2 are arbitrary constants, ξ and η satisfy (7) and (8), respectively.

3.2. The double non-traveling wave solutions of (2 + 1)-dimensional breaking soliton equation

Now, we consider (2 + 1)-dimensional BS equation (2) and search for its double non-traveling wave solutions. By
balancing the highest order derivative terms and nonlinear terms in Eq. (2), we can write the solutions of Eq. (2) in the form
  ′  ′  ′ 2  ′  ′   ′ 2
G1 G2 G1 G1 G2 G2
,

u = a0 + a10 + a01 + a20 + a11 + a02


G1 G2 G1 G1 G2 G2
2 2 (23)
′ ′ ′ ′ ′
G′2
        
G1 G2 G1 G1 G2
v = b0 + b10 ,

+ b01 + b20 + b11 + b02


G1 G2 G1 G1 G2 G2

where G1 = G1 (ξ ), G2 = G2 (η) satisfy Eq. (5), and ξ , η, a0 , a10 , a01 , a20 , a11 , a02 , b0 , b10 , b01 , b20 , b11 , b02 are functions
G′ G′
of (x, y, t ) to be determined later. Using Eq. (5), substituting (23) into Eq. (2) and equating the coefficients of ( G1 )i ( G2 )j to
1 2
 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816 811

zero, we obtain a set of over-determined partial equations for ξ , η, a0 , a10 , a01 , a20 , a11 , a02 , b0 , b10 , b01 , b20 , b11 , b02 .
Solving the equations, we obtain

C1 φ(t ) + (C4 t + C5 )C2 − C1 C3 y

ξ = C1 x + , η = φ(t ) + C2 x + C3 y, (24)
C2

(−βλ21 − 8βµ1 )C13 C3 + (−βλ22 − 8βµ2 )C1 C22 C3 + C2 C4

a0 = , (25)
8β C1 C3

3C12 λ1 3C22 λ2 3C12 3C22

a10 = − , a01 = − , a20 = − , a11 = 0, a02 = − , (26)
2 2 2 2
(βλ21 + 8βµ1 )C13 C3 + [(−βλ22 − 8βµ2 )C22 C3 − 2φ ′ (t )]C1 − C2 C4
b0 = , (27)
8 β C1 C2

3λ1 C12 C3 3λ2 C2 C3 3C12 C3 3C2 C3

b10 = , b01 = − , b20 = , b11 = 0, b02 = − , (28)
2C2 2 2C2 2

where φ(t ) is an arbitrary function of t, C1 , C2 , C3 , C4 , and C5 are arbitrary constants. Substituting (24)–(28) into (23), we
can rewrite the solutions of Eq. (2) in the following form

3C12 λ1 G′1 3C22 λ2 G′2

  2  2
3C12 G′1 3C22 G′2
   
,


 u = a 0 − − − −
2 G1 2 G2 2 G1 2 G2

(29)
3λ C 2 C 3λ2 C2 C3 G′2
 ′  2  2
3C 2 C3 G′1 3C2 C3 G′2
 
G1
v = b0 + 1 1 3

+ 1 .

 − −
2C2 G1 2 G2 2C2 G1 2 G2

By using the general solutions of LODEs (5), we obtain the following double non-traveling wave solutions of Eq. (2).

λ21 − 4µ1 > 0
Case 1: When , we obtain the double hyperbolic function solution
λ22 − 4µ2 > 0

        2
A sinh 1
λ 2
− 4 µ ξ + A cosh 1
λ 2
− 4µ ξ

1 1 2 1
 1 1
3C12 2 2 2


u1 (x, y, t ) = − 8 (λ1 − 4µ1 ) 

  
       
A1 cosh 12 λ21 − 4µ1 ξ + A2 sinh 12 λ21 − 4µ1 ξ









        2
1
λ 2
µ η 1
λ 2
µ η

 B sinh − 4 + B cosh − 4
 1 2 2 2 2 2

3C 2 2 2

 − 2 (λ22 − 4µ2 ) 

 
    
8
    
B1 cosh 21 λ22 − 4µ2 η + B2 sinh 21 λ22 − 4µ2 η







3λ2 C 2 3λ2 C 2


 + a0 + 1 1 + 2 2 ,



8 8
       2 (30)
λ µ ξ λ µ ξ
 1 2 1 2
A sinh − 4 + A cosh − 4

1 1 2 1

1 1
3C12 C3 2 2 2


v1 (x, y, t ) = (λ1 − 4µ1 ) 

  
      
8C2  
A1 cosh 21 λ21 − 4µ1 ξ + A2 sinh 12 λ21 − 4µ1 ξ









        2
1
λ 2
µ η 1
λ 2
µ η

 B sinh − 4 + B cosh − 4
 1 2 2 2
 2 2
2 2
3C2 C3 2


(λ2 − 4µ2 ) 
 
 −
    
8
    
B1 cosh 12 λ22 − 4µ2 η + B2 sinh 21 λ22 − 4µ2 η







 + b0 − 3λ1 C1 C3 + 3λ2 C2 C3 ,
2 2 2




8C2 8

C φ(t )+(C t +C )C −C C y
where ξ = C1 x + 1 4
C2
5 2 1 3
, η = φ(t ) + C2 x + C3 y, φ(t ) is an arbitrary function of t, a0 and b0 satisfy (25) and
(27) respectively, and A1 , A2 , B1 , B2 , C1 , C2 , C3 , C4 , C5 are arbitrary constants.
812 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816

Setting A1 , B1 > 0, A21 > A22 , B21 > B22 in the solution (30), we can rewrite u1 (x, y, t ) and v1 (x, y, t ) as
3C12 2
  
 1
u11 (x, y, t ) = 8 (λ1 − 4µ1 ) sech 2 λ1 − 4µ1 ξ + ξ0
 2 2






3µ1 C12 3µ2 C22

3C 2
  
1


 + 2 (λ22 − 4µ2 ) sech2 λ22 − 4µ2 η + η0 + a0 + ,

 +
8 2 2 2

(31)
3C12 C3 2
 
1

v ( , , ) (λ µ ) 2
λ21 − 4µ1 ξ + ξ0


 11 x y t = − 1 − 4 1 sech
8C2 2







 + 3C2 C3 (λ2 − 4µ ) sech2 1 λ2 − 4µ η + η + b − 3µ1 C1 C3 + 3µ2 C2 C3 ,

2

   

2 2 2 2 0 0
8 2 2C2 2
which is the double solitary-like wave solution of the (2 + 1)-dimensional BS equation (2). Here, ξ0 = tanh−1 ( A2 ), η0 =
A
1
tanh−1 ( B2 ).
B
1
In particular, setting A1 ̸= 0, B1 ̸= 0, A2 = B2 = 0 in the solution (30), we obtain the following double solitary-like wave
solution of Eq. (2)
3C12 2 3C22 2
     
1 1

u12 (x, y, t ) = − 8 (λ1 − 4µ1 ) tanh 2 λ1 − 4µ1 ξ − 8 (λ2 − 4µ2 ) tanh 2 λ2 − 4µ2 η
 2 2 2 2



3λ2 C 2 3λ2 C 2


 + a0 + 1 1 + 2 2 ,



8 8
(32)
3C12 C3 2
     
1 3C2 C3 2 1
v ( , , ) (λ µ ) 2
λ 2
µ ξ (λ µ ) 2
λ2
µ η


 12 x y t = 1 − 4 1 tanh 1 − 4 1 − 2 − 4 2 tanh 2 − 4 2
8C2 2 8 2




 + b − 3λ1 C1 C3 + 3λ2 C2 C3 .

 2 2 2

0
8C2 8
In view of the above exact solution (32), the arbitrary function φ(t ) is the main factor having effect on dynamic properties
of double solitary-like waves. In Fig. 3, we present the profiles of the double solitary-like waves for fixed time. We can see
from plot (a) that u12 is combined with two single bright solitary-like waves under certain conditions to form bright–bright
solitary-like waves, while plot (b) shows that v12 is composed of bright and dark solitary-like waves which can propagate
simultaneously in the nonlinear medium and sit on the same background. Fig. 4 shows the dynamic properties of the double
solitary-like wave solution expressed by Eq. (32), i.e., the interactions between the solitary-like waves. Through the studies
on plots (a) and (b) in Fig. 4, we can see that u12 and v12 maintain their original dynamic properties including the wave shape
and amplitude after the collisions, which indicates that the progress is elastic interaction.
When setting A1 , A2 , B1 , B2 , λ1 , λ2 , µ1 , µ2 , and φ(t ) to other special values in a proper way, we can obtain more double
solitary-like wavesolutions of Eq. (2), and here we do not list all of them.
λ21 − 4µ1 < 0
Case 2: When , we obtain the double trigonometric function solution
λ22 − 4µ2 < 0
        2
A cos 2 4µ1 − λ1 ξ − A2 sin 2 4µ1 − λ1 ξ
1 2 1 2
 1

3C 2


u2 (x, y, t ) = − 1 (4µ1 − λ21 ) 

 
      
8
  
A1 sin 21 4µ1 − λ21 ξ + A2 cos 21 4µ1 − λ21 ξ







       2

 
B cos 2 4µ2 − λ2 η − B2 sin 2 4µ2 − λ2 η
1 2 1 2


3C22  1 3λ2 C 2 3λ2 C 2


( µ λ ) + a0 + 1 1 + 2 2 ,
 2 

− 4 −

2 

2 
     
 8  8 8
B1 sin 12 4µ2 − λ22 η + B2 cos 12 4µ2 − λ22 η




       2 (33)
µ λ ξ µ λ ξ
 1 1
2 2
A cos 4 − − A sin 4 −


3C12 C3  1 2 1 1 2 2 1 1


v (x, y, t ) = (4µ1 − λ21 ) 

 
2 
   
8C

    
2
A1 sin 12 4µ1 − λ21 ξ + A2 cos 21 4µ1 − λ21 ξ







        2

B cos 2 4µ2 − λ2 η − B2 sin 2 4µ2 − λ2 η
1 2 1 2


 1
 + b0 − 3λ1 C1 C3 + 3λ2 C2 C3 ,
 2 2 2
3C2 C3

( µ λ )
 2 

− 4 −


 2 2
   
8 8C2 8
    
B1 sin 12 4µ2 − λ22 η + B2 cos 12 4µ2 − λ22 η


 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816 813

Fig. 3. The double solitary-like wave solutions for the fields u11 (a) and v11 (b) expressed by Eq. (32) with C1 = C2 = C3 = C4 = C5 = β = 1, λ1 = 1,
λ2 = 2, µ1 = 0.1, µ2 = 0.5, φ(t ) = 1 at the fixed time t = 10.

Fig. 4. The evolutional plots of double solitary-like structures u11 (a) and v11 (b) expressed by Eq. (32) with the same parameters as in Fig. 3.

C φ(t )+(C t +C )C −C C y
where ξ = C1 x + 1 4
C2
5 2 1 3
, η = φ(t ) + C2 x + C3 y, φ(t ) is an arbitrary function of t, a0 and b0 satisfy (25) and
and A1 , A2 , B1 , B2 , C1 , C2 , C3 , C4 , C5 are arbitrary constants.
(27) respectively, 
λ21 − 4µ1 > 0
Case 3: When , we get the complexiton solution combining hyperbolic and trigonometric function
λ22 − 4µ2 < 0

        2
A sinh 2 λ1 − 4µ1 ξ + A2 cosh 2 λ1 − 4µ1 ξ
1 2 1 2
 1

3C12 2


( , , ) (λ µ )
 

 u 3 x y t = − 1 − 4 1
      
8
  
A1 cosh 12 λ21 − 4µ1 ξ + A2 sinh 12 λ21 − 4µ1 ξ







       2

 
1
µ λ 2
η 1
µ λ 2
η

 B1 cos 4 2 − − B2 sin 4 2 −
3C 2 2 2
3λ2 C 2 3λ2 C 2

 2 2
− 2 (4µ2 − λ22 )  + a0 + 1 1 + 2 2 ,

  

      
 8   8 8
B1 sin 12 4µ2 − λ22 η + B2 cos 12 4µ2 − λ22 η




       2 (34)
λ µ ξ λ µ ξ
 1 1
2 2
A sinh − 4 + A cosh − 4


3C12 C3 2  1 2 1 1 2 2 1 1


v ( x, y, t ) = (λ 4 µ1 ) 
 

 3 1 −    
8C

    
2
A1 cosh 21 λ21 − 4µ1 ξ + A2 sinh 12 λ21 − 4µ1 ξ







        2

B cos 2 4µ2 − λ2 η − B2 sin 2 4µ2 − λ2 η
1 2 1 2


 1
 + b0 − 3λ1 C1 C3 + 3λ2 C2 C3 ,
 2 2 2
3C2 C3

( µ λ )
 2 

− 4 −


 2 2 
   
8 8C2 8
   
B sin 1 4µ − λ2 η + B cos 1 4µ − λ2 η


1 2 2 2 2 2 2 2

C φ(t )+(C t +C )C −C C y
where ξ = C1 x + 1 4
C2
5 2 1 3
, η = φ(t ) + C2 x + C3 y, φ(t ) is an arbitrary function of t, a0 and b0 satisfy (25) and
(27) respectively, and A1 , A2 , B1 , B2 , C1 , C2 , C3 , C4 , C5 are arbitrary constants.
814 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816

λ21 − 4µ1 < 0
Case 4: When , we obtain the complexiton solution combining hyperbolic and trigonometric function
λ22 − 4µ2 > 0

        2
A cos 2 4µ1 − λ1 ξ − A2 sin 2 4µ1 − λ1 ξ
1 2 1 2
 1

3C12


( , , ) ( µ λ )
 2 


 u 4 x y t = − 4 1 − 1 
     
8
 
A1 sin 21 4µ1 − λ21 ξ + A2 cos 21 4µ1 − λ21 ξ







       2

 
1
λ 2
µ η 1
λ 2
µ η

 B 1 sinh − 4 2 + B2 cosh − 4 2
3C 2 2 2
3λ2 C 2 3λ2 C 2

 2 2
− 2 (λ22 − 4µ2 )  + a0 + 1 1 + 2 2 ,

  

      
 8   8 8
B1 cosh 21 λ22 − 4µ2 η + B2 sinh 21 λ22 − 4µ2 η




       2 (35)
A cos 2 4µ1 − λ1 ξ − A2 sin 2 4µ1 − λ1 ξ
 1 1

 2 2
3C12 C3  1


v (x, y, t ) = (4µ1 − λ21 ) 

 
4 
     
8C

  
2
A1 sin 12 4µ1 − λ21 ξ + A2 cos 21 4µ1 − λ21 ξ







        2

B sinh 2 λ2 − 4µ2 η + B2 cosh 2 λ2 − 4µ2 η
1 2 1 2


 1
 + b0 − 3λ1 C1 C3 + 3λ2 C2 C3 ,
 2 2 2
3C2 C3 2

(λ µ )
 
− − 4

 2

2
    
8 8C2 8
    
B cosh 1 λ2 − 4µ η + B sinh 1 λ2 − 4µ η


1 2 2 2 2 2 2 2

C φ(t )+(C t +C )C −C C y
where ξ = C1 x + 1 4
C2
5 2 1 3
, η = φ(t ) + C2 x + C3 y, φ(t ) is an arbitrary function of t, a0 and b0 satisfy (25) and
1 , A2 , B1 , B2 , C
(27) respectively, and A 1 , C2 , C3 , C4 , C5 are arbitraryconstants. 
λ21 − 4µ1 > 0 λ21 − 4µ1 = 0 λ21 − 4µ1 < 0 λ21 − 4µ1 = 0 λ21 − 4µ1 = 0
In other cases, i.e., , , , , , we obtain combining
λ22 − 4µ2 = 0 λ22 − 4µ2 > 0 λ22 − 4µ2 = 0 λ22 − 4µ2 < 0 λ22 − 4µ2 = 0
hyperbolic and rational function solution, combining trigonometric and rational function solution and double rational
solution of Eq. (2) by the same procedure as in Section 3.1. For simplicity, we do not list all of them here.

Remark 1. All the double non-traveling wave solutions presented in this paper have been verified with Maple package
′
by putting them back into the original Eqs. (1) and (2), respectively. The above examples show that the compound ( GG )-
expansion method can be easily used to obtain more exact solutions of NPDEs with free parameters A1 , A2 , B1 , B2 , λ1 , λ2 , µ1
and µ2 .
′
Remark 2. For the second order ordinary equation G′′ + aG′ + bG = 0, the transformation f = ( GG ) satisfies the Riccati
equation dξ = −b − af − f 2 . Here, G = G(ξ ), a and b are constants. Therefore, the present method is also an application
df

of the expansion method in Ref. [30]. The solutions of Riccati equation published in Ref. [30], including hyperbolic function
solutions, trigonometric function solutions, and rational solutions, can also be used to construct exact solutions of nonlinear
partial differential equations.
′
Remark 3. The ansätz (4) proposed here is more general than the ansätz in ( GG )-expansion method [25] and the generalized
G′
( G )-expansion method [26]. If we set the parameters in (4) and (5) to different values, the above two methods can be
recovered by our proposed method.

Remark 4. We assume that the coefficients in ansätz (4) are the functions of (x1 , x2 , . . . , t ) and we introduce two indepen-
dent variables ξ and η, which are different and not proportional to each other. Therefore, it is possible to obtain double non-
traveling wave solutions including double hyperbolic function solutions, double trigonometric function solutions, double
′
rational solutions and complexiton solutions of NPDEs by the compound ( GG )-expansion method. By setting the parameters
to special values in a proper way, we can obtain the double solitary-like wave solutions. However, there exists only one
′ ′
independent variable in ( GG )-expansion method [25] and the generalized ( GG )-expansion method [26], which indicates that
the double solitary-like wave solutions and complexiton solutions cannot be obtained by using the above two methods.

Remark 5. We can further extend our method by introducing a more generalized ansätz as follows:

G′1 (ξ1 ) G′2 (ξ2 ) G′m (ξm )

n
   r1  r2  rm
u(x1 , x2 , . . . , t ) = a0 + ar1 r2 ···rm ···
k=1 r1 +r2 +···+rm =k
G1 (ξ1 ) G2 (ξ2 ) Gm (ξm )

G′1 (ξ1 ) G′2 (ξ2 ) G′m (ξm )

−1
   s1  s2  sm
+ bs1 s2 ···sm ··· , (36)
l=−n s1 +s2 +···+sm =l
G1 (ξ1 ) G2 (ξ2 ) Gm (ξm )
 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816 815

where a0 , ar1 r2 ···rm , bs1 s2 ···sm and ξi (i = 1, 2, . . . , m) are functions of (x1 , x2 , . . . , t ) to be determined, the positive integer
n can be determined by balancing the highest order derivatives and nonlinear terms in the given NPDEs, ri > 0, si 6 0 (i =
1, 2, . . . , m) and Gi = Gi (ξi ) (i = 1, 2, . . . , m) satisfy the following second order LODEs

G′′i + λi G′i + µi Gi = 0, i = 1, 2, . . . , m (37)

where λi , µi are constants, and i̸=j [(λi −λj ) +(µi −µj ) ] ̸= 0 (i, j = 1, 2, . . . , m). By using ansätz (36), it is clear that we
2 2

can get more novel multiple non-traveling wave solutions, multiple soliton solutions, and complexiton solutions of NPDEs.

4. Conclusion
′ ′
As an extension of the ( GG )-expansion method, the compound ( GG )-expansion method is proposed to construct the mul-
tiple non-traveling wave solutions of NPDEs in this work. As applications of the proposed method, we obtain a series of
double non-traveling wave solutions of (2 + 1)-dimensional Painlevé integrable Burgers equation and (2 + 1)-dimensional
breaking soliton equation. These solutions include double hyperbolic function solutions, double trigonometric function solu-
tions, double rational solutions, and complexiton solutions. Such complexiton solutions possess combination of hyperbolic
and trigonometric function solutions, combination of hyperbolic and rational function solutions, combination of trigono-
metric and rational function solutions. These exact solutions involve the arbitrary functions, which is appropriate to excite
rich nonlinear structures. When the parameters are taken as special values, the double solitary-like wave solutions are de-
rived from the double hyperbolic function solutions. In addition, we find that the double solitary-like wave solutions of
(2 + 1)-dimensional PIB equation are composed of two single kink solitary solutions and have the same shock font. The
(2 + 1)-dimensional BS equation can admit the bright–bright and bright–dark solitary-like wave pairs. Moreover, the soli-
tonic interactions are investigated and we find that the collision between the solitary-like waves is elastic. The Remarks 3
′
and 4 illustrate that the compound ( GG )-expansion method is more powerful than the methods in Refs. [25,26]. The present
′
work shows that the compound ( GG )-expansion method is direct, effective and can be applied to many other NPDEs in
mathematical physics, such as the approximate long water wave equation [31,32], the variant Boussinesq equation [33],
the (2 + 1)-dimensional Boiti–Leon–Pempinelli(BLP) equation [34], the coupled Whitham–Broer–Kaup equation [35], and
Klein–Gordon equation [36].

Acknowledgments

The authors express their sincere thanks to the Editor and referees for their valuable comments which led to an improved
version.

References

[1] Deng-Shan Wang, Hongbo Li, Symbolic computation and non-travelling wave solutions of (2 + 1)-dimensional nonlinear evolution equations, Chaos
Solitons Fractals 38 (2008) 383–390.
[2] Bao-Dong Wang, Li-Na Song, Hong-Qing Zhang, A new extended elliptic equation rational expansion method and its application to (2 + 1)-dimensional
Burgers equation, Chaos Solitons Fractals 33 (2007) 1546–1551.
[3] Sheng Zhang, A further improved extended Fan sub-equation method for (2 + 1)-dimensional breaking soliton equations, Appl. Math. Comput. 199
(2008) 259–267.
[4] Sheng Zhang, A generalized new auxiliary equation method and its application to the (2 + 1)-dimensional breaking soliton equations, Appl. Math.
Comput. 190 (2007) 510–516.
[5] M.J. Ablowitz, P.A. Clarkson, Soliton, Nonlinear Evolution Equations and Inverse Scattering, Cambridge Univ. Press, New York, 1991.
[6] F. Cariello, M. Tabor, Similarity reductions from extended Painlev expansions for nonintegrable evolution equations, Physica D 53 (1991) 59–70.
[7] C. Rogers, W.F. Shadwick, Bäcklund Transformations, Academic Press, New York, 1982.
[8] V.A. Matveev, M.A. Salle, Darboux Transformation and Solitons, Springer, Berlin, 1991.
[9] P.J. Olver, Applications of Lie Groups to Differential Equations, second ed., Springer Verlag, New York, 1993.
[10] Mingliang Wang, Solitary wave solutions for variant Boussinesq equations, Phys. Lett. A 199 (1995) 169–172.
[11] Yubin Zhou, Mingliang Wang, Tiande Miao, The periodic wave solutions and solitary wave solutions for a class of nonlinear partial diffrential equations,
Phys. Lett. A 323 (2004) 77–88.
[12] Ji-Huan He, Xu-Hong Wu, Exp-function method for nonlinear wave equations, Chaos Solitons Fractals 30 (2006) 700–708.
[13] A. Ebaid, Exact solitary wave solutions for some nonlinear evolution equations via Exp-function method, Phys. Lett. A 365 (2007) 213–219.
[14] Shimin Guo, Yubin Zhou, Auxiliary equation method for the mKdV equation with variable coefficients, Appl. Math. Comput. 217 (2010) 1476–1483.
[15] Li Zhang, Yezhi Lin, Yinping Liu, New solitary wave solutions for two nonlinear evolution equations, Comput. Math. Appl. 67 (2014) 1595–1606.
[16] Hong Li, Kanmin Wang, Jibin Li, Exact traveling wave solutions for the Benjamin–Bona–Mahony equation by improved Fan sub-equation method,
Appl. Math. Model. 37 (2013) 7644–7652.
[17] Engui Fan, Extended tanh-function method and its applications to nonlinear equations, Phys. Lett. A 277 (2000) 212–218.
[18] Engui Fan, Jian Zhang, Applications of the Jacobi elliptic function method to special-type nonlinear equations, Phys. Lett. A 305 (2002) 383–392.
[19] Abdul-Majid Wazwaz, A sine–cosine method for handlingnonlinear wave equations, Math. Comput. Modelling 40 (2004) 499–508.
[20] Abdul-Majid Wazwaz, Kink solutions for three new fifth order nonlinear equations, Appl. Math. Model. 38 (2014) 110–118.
[21] Z.Z. Ganji, D.D. Ganji, H. Bararnia, Approximate general and explicit solutions of nonlinear BBMB equations by Exp-function method, Appl. Math.
Model. 33 (2009) 1836–1841.
[22] Huiqun Zhang, Wen-Xiu Ma, Extended transformed rational function method and applications to complexiton solutions, Appl. Math. Comput. 230
(2014) 509–515.
[23] Wen-Xiu Ma, Zuonong Zhu, Solving the (3+1)-dimensional generalized KP and BKP equations by the multiple exp-function algorithm, Appl. Math.
Comput. 218 (2012) 11871–11879.
816 S. Guo et al. / Computers and Mathematics with Applications 69 (2015) 804–816

[24] Wen-Xiu Ma, Tingwen Huang, Yi Zhang, A multiple exp-function method for nonlinear differential equations and its application, Phys. Scr. 82 (2010)
065003.
′
[25] Mingliang Wang, Xiangzheng Li, Jinliang Zhang, The ( GG )-expansion method and travelling wave solutions of nonlinear evolution equations in
mathematical physics, Phys. Lett. A 372 (2008) 417–423.
′
[26] Sheng Zhang, Jing-Lin Tong, Wei Wang, A generalized ( GG )-expansion method for the mKdV equation with variable coeffcients, Phys. Lett. A 372
(2008) 2254–2257.
′
[27] Shimin Guo, Yubin Zhou, The extended ( GG )-expansion method and its applications to the Whitham–Broer–Kaup–Like equations and coupled
Hirota–Satsuma KdV equations, Appl. Math. Comput. 215 (2010) 3214–3221.
′
[28] Shimin Guo, Yubin Zhou, Chenxia Zhao, The improved ( GG )-expansion method and its applications to the Broer–Kaup equations and approximate
long water wave equations, Appl. Math. Comput. 216 (2010) 1965–1971.
[29] Wen-Xiu Ma, Complexiton solutions to the Korteweg–de Vries equation, Phys. Lett. A 301 (2002) 35–44.
[30] W.X. Ma, B. Fuchssteiner, Explicit and exact solutions to a Kolmogorov–Petrovskii–Piskunov equation, Int. J. Non-Linear Mech. 31 (1996) 329–338.
′
[31] Mingliang Wang, Jinliang Zhang, Xiangzheng Li, Application of the ( GG )-expansion to travelling wave solutions of the Broer–Kaup and the approximate
long water wave equations, Appl. Math. Comput. 206 (2008) 321–326.
[32] Yong Chen, Qi Wang, A new elliptic equation rational expansion method and its application to the shallow long wave approximate equations, Appl.
Math. Comput. 173 (2006) 1163–1182.
[33] Emmanuel Yomba, The extended Fans sub-equation method and its application to KdV-MKdV, BKK and variant Boussinesq equations, Phys. Lett. A
336 (2005) 463–476.
[34] Chaoqing Dai, Yueyue Wang, Periodic structures based on variable separation solution of the (2 + 1)-dimensional Boiti–Leon–Pempinelli equation,
Chaos Solitons Fractals 39 (2009) 350–355.
[35] A.A. Imani, D.D. Ganji, Houman B. Rokni, et al., Approximate traveling wave solution for shallow water wave equation, Appl. Math. Model. 36 (2012)
1550–1557.
[36] Zaid Odibat, Construction of solitary solutions for nonlinear dispersive equations by variational iteration method, Phys. Lett. A 372 (2008) 4045–4052.