Dear Author, Please, note that changes made to the HTML content will be added to the article before publication, but are not reflected in this PDF. Note also that this file should not be used for submitting corrections. ARTICLE IN PRESS JID: CHAOS [m3Gdc;February 24, 2016;10:55] Chaos, Solitons and Fractals xxx (2016) xxx–xxx Contents lists available at ScienceDirect Chaos, Solitons and Fractals Nonlinear Science, and Nonequilibrium and Complex Phenomena journal homepage: www.elsevier.com/locate/chaos Group analysis, exact solutions and conservation laws of a generalized fifth order KdV equation Gangwei Wang a,∗, Abdul H. Kara b, Kamran Fakhar c, Jose Vega-Guzman d, Anjan Biswas e,f Q1 Q2 a School of Mathematics and Statistics, Beijing Institute of Technology, Beijing 100081, PR China School of Mathematics, Centre for Differential Equations Continuum Mechanics and Applications, University of the Witwatersrand, Johannesburg Wits 2050, South Africa c Department of Mathematics, University of British Columbia, Vancouver V6T 1Z2, Canada d Department of Mathematics, Howard University, Washington, DC 20059, USA e Department of Mathematical Sciences, Delaware State University, Dover, DE 19901-2277, USA f Department of Mathematics, Faculty of Science, King Abdulaziz University, Jeddah 21589, Saudi Arabia b a r t i c l e i n f o Article history: Received 22 January 2016 Accepted 11 February 2016 Available online xxx PACS: 02.30.Jr 05.45.Yv 02.30.Ik a b s t r a c t We study the generalized fifth order KdV equation using group methods and conservation laws. All of the geometric vector fields of the special fifth order KdV equation are presented. By using the nonclassical Lie group method, it is show that this equation does not admit nonclassical type symmetries. Then, on the basis of the optimal system, the symmetry reductions and exact solutions to this equation are constructed. For some special cases, we obtain additional nontrivial conservation laws and scaling symmetries. © 2016 Published by Elsevier Ltd. Keywords: Fifth order KdV equation Symmetry analysis Exact solutions Conservation laws 1 2 3 4 5 6 7 8 9 10 1. Introduction Nonlinear evolution equations (NLEEs) are of importance in nonlinear science, in particular in applied mathematics and theoretical physics. Their solutions are important in the understanding of nonlinear interaction and behaviors of complex system. There are various techniques [1–17] used to deal with NLEEs, some of the commonly used ways involve the generalized symmetries, nonlocal symmetries, nonclassical Lie group and classical Lie group method. It is well known that differential equations (DEs) have a number fundamental structures, that is, symmetries and conservation laws (CLs). CLs play a key roles in DEs analysis, particularly in studies of existence, uniqueness and stability of solutions. Various approaches have been used to handle symmetries and conservation laws of PDE systems (see [1–6] and the references therein). In the present paper, we use the group method and the multiplier approach to study the fifth order KdV equation ut + α un ux + β ux uxx + γ uuxxx + uxxxxx = 0. Corresponding author. Tel.: +86 13683181850. E-mail address: wanggangwei@bit.edu.cn (G. Wang). ut + α u2 ux + β ux uxx + γ uuxxx + uxxxxx = 0, 13 14 15 16 17 18 19 20 (1) In particular, for n = 2, one can get ∗ 11 12 21 (2) http://dx.doi.org/10.1016/j.chaos.2016.02.013 0960-0779/© 2016 Published by Elsevier Ltd. Please cite this article as: G. Wang et al., Group analysis, exact solutions and conservation laws of a generalized fifth order KdV equation, Chaos, Solitons and Fractals (2016), http://dx.doi.org/10.1016/j.chaos.2016.02.013 JID: CHAOS 2 22 23 24 25 26 ARTICLE IN PRESS G. Wang et al. / Chaos, Solitons and Fractals xxx (2016) xxx–xxx here α , β and γ are nonzero constants. Clearly, this equation has two dispersive terms uxxx and uxxxxx . By choosing the real values of the parameters n, α , β and γ , one can get a variety of fKdV equations [7,8] such as, when n = 2, the Sawada–Kotera (SK) equation ut + 5u2 ux + 5ux uxx + 5uuxxx + uxxxxx = 0, 27 29 ut + 2u ux + 6ux uxx + 3uuxxx + uxxxxx = 0. 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 (4) (6) (7) Many authors have been studied these equations using different approaches. There is still, however, a lot of room for extensions and improvements. in particular exact solutions, symmetries and conservation law. The main purpose of this paper is to investigate symmetry and conservation law classification of the generalized KdV equation. We will show that the particular case n = 2 is special as it is the only case that admit a scaling symmetry. The paper is organized as follows. In Section 2, all of the vector fields of Eq. (1) are constructed. In Section 3, we consider the special case of n, that is n = 2. In this case, the vector fields and some exact solutions are obtained. In Section 4, we study the soliton solutions of the equation. In Section 5, we find the conservation laws. Finally, conclusions and some remarks are given in Section 5. 2. Group analysis of the generalized fifth-order KdV equation Consider a one-parameter Lie group of infinitesimal transformation: x∗ = x + ξ (x, t, u ) + O( 2 ), u∗ = u + η (x, t, u ) + O( 2 ), 51 52 with a small parameter   1, and the above group of transformations infinitesimal generator can read ∂ ∂ ∂ V = τ (x, t, u ) + ξ (x, t, u ) + η (x, t, u ) , ∂t ∂x ∂u 53 54 55 (9) and we need to solve the coefficient functions τ (x, t, u), ξ (x, t, u), η(x, t, u). Meanwhile, V must satisfy Lie’s symmetry condition pr 56 (8) (5 ) V ( )| =0 = 0, (10) where = ut + α u2 ux + β ux uxx + γ uuxxx + uxxxxx . (11) 59 60 (12) 61 η = Dt (η ) − ux Dt (ξ ) − ut Dt (τ ), ηx = Dx (η ) − ux Dx (ξ ) − ut Dx (τ ), (13) Here, Di are the total derivative operators defined by Di = ∂ ∂ ∂ + ui + ui j + · · · i = 1, 2, ∂u ∂uj ∂ xi and (x1 , x2 ) = (t, x ). On the basis of the Lie symmetry analysis method, one can get τ = c1 , ξ = c2 , η = 0, ∂ ∂ ,V = . ∂ x 2 ∂t 63 64 65 (15) where c1 and c2 are arbitrary constants. So one can have the geometric vector fields V1 = 62 (14) 66 67 (16) 3. Classical, nonclassical, potential symmetry and exact solutions for n = 2 68 In this section, we use the classical and nonclassical symmetry method to handle the fifth-order equation for n = 2. 70 3.1. Classical symmetry analysis 73 On the basis of the Lie symmetry analysis method, one can get 75 τ = 5c1t + c2 , ξ = c1 x + c3 , η = −2c1 u, V1 = ∂ ∂ ∂ ∂ ∂ ,V = , V =x + 5t − 2u . ∂ x 2 ∂t 3 ∂x ∂t ∂u 69 71 72 74 (17) where c1 , c2 and c3 are arbitrary constants. So one can have the geometric vector fields t ∗ = t + τ (x, t, u ) + O( 2 ), 58 t ··· (5) the Ito equation 2 31 ηt + nαηux + α un ηx + βηx uxx + βηxx ux +γ ηuxxx + γ ηxxx u + ηxxxxx = 0, the Kaup–Kupershmidt (KK) equation ut + 20u2 ux + 25ux uxx + 10uuxxx + uxxxxx = 0, 30 By using the fifth prolongation Pr(5) V to Eq. (1), one can see that the coefficient functions satisfy the following equation: where the Lax equation ut + 30u2 ux + 20ux uxx + 10uuxxx + uxxxxx = 0, Q3 (3) the Caudrey–Dodd–Gibbon (CDG) equation ut + 180u2 ux + 30ux uxx + 30uuxxx + uxxxxx = 0, 28 [m3Gdc;February 24, 2016;10:55] 76 77 (18) 3.2. Nonclassical symmetry analysis 78 In the previous subsection, we used the classical symmetry method to deal with the fifth order KdV equation. Next, we employ the nonclassical symmetry method [9,10] to study the fifth order KdV equation. The aim is that nonclassical symmetries are much more numerous than classical ones and maybe get more solutions of PDEs. In terms of the classical symmetry, the invariance surface condition should been added: 79 1 = η − ξ ux − τ ut . 80 81 82 83 84 85 86 (19) 57 87 Please cite this article as: G. Wang et al., Group analysis, exact solutions and conservation laws of a generalized fifth order KdV equation, Chaos, Solitons and Fractals (2016), http://dx.doi.org/10.1016/j.chaos.2016.02.013 JID: CHAOS ARTICLE IN PRESS [m3Gdc;February 24, 2016;10:55] G. Wang et al. / Chaos, Solitons and Fractals xxx (2016) xxx–xxx 88 89 If the vector field (9) is a nonclassical symmetry of (1), which should satisfy pr (5)V ( )| =0, 1 =0 = 0. 90 91 92 93 94 95 96 97 Consider the nature of the invariant surface condition (19), without loss of generality, there are two cases to arise: (i)τ = 1; (ii)τ = 0, ξ = 1. More details see [9,10] and the references therein. In following, we will consider them respectively. 1. τ = 1. Then from the invariance surface condition (19), one can get ut = η − ξ ux . 98 99 (20) (21) After differentiating (21) with respect to x, and then replacing ut by using η − ξ ux , one can get ηt − ξt ux + ηu (η − ξ ux ) + 2αηuux + α u2 (ηx + (ηu − ξx )ux ) +β uxx (ηx + (ηu − ξx )ux ) + β ux (ηxx + (2ηxu − ξxx )ux +(ηuu −2ξxu u2x ) + (ηu −2ξx uxx )) + γ ηuxxx +γ u[ηxxx + 3ηxxu ux + 3ηxuu u2x + 3ηxu uxx + ηuuu u3x + 3ηuu ux uxx +ηu uxxx − ξxxx ux −3ξxx uxx −3ξx uxxx ] + 5ηxxxxu ux + 10ηxxxu uxx +10ηxxu uxxx + 3ηxu uxxxx + ηu uxxxxx − ξxxxxx ux −5ξxxxx uxx −10ξxxx uxxx +10ηuu uxx uxxx + ηxxxxx − 10ξxx uxxxx −5ξx uxxxxx +10ηxxuu u2x +10ηxxuuu u3x + 5ηxuuuu u4x + 15ηxuuu u2xx + ηuuuuu u5x +30ηxuuu u2x uxx + 20ηxuu ux uxxx + 30ηxxuu ux uxx + 5ηuu ux uxxxx +15ηuuu ux u2xx +10ηuuuu u3x uxx +10ηuuu u2x uxxx = 0. (22) 100 102 104 105 106 107 108 −2u x + C2 ,ξ = , 5t + C1 5t + C1 x + C2 ∂ ∂ −2u ∂ + + . 5t + C1 ∂ x ∂ t 5t + C1 ∂ u (24) It is cleat that, V4 = V3 + C1V2 + C2V1 , it is the classical symmetry. Also, we can find the equation does not have nonclassical symmetries. 2. τ = 0, ξ = 1. Now, using the same approach as before, one can have η = ux . 109 (23) where C1 and C2 are arbitrary constants. Consequently, one can get the corresponding “nonclassical symmetry”is V4 = 103 (26) 113 That is to say, in this case, we could not get supplementary symmetries, of non-classical type. This also means no new explicit solutions can be constructed in the case of τ = 0, ξ = 1. 114 3.3. Potential symmetry analysis 111 112 115 vx = u, 1  1 vt = − α u3 + (β − γ )u2x + γ uuxx + uxxxx . 3 2 (28) After repeating previous steps, one can get the coefficient functions τ (x, t, u, v ), ξ (x, t, u, v ), η (x, t, u, v ) and ψ (x, t, u, v ) are: τ = c1 t + c2 , ξ = 118 119 120 2 c1 c1 x + c4 , η = − u, ψ = −c1 v + c3 . 3 3 (29) One can find out that this equation does not have potential symmetry. 121 122 3.4. Symmetry reductions and group-invariant solutions for n=2 123 In the previous section, we use the classical and nonclassical group method to deal with the fifth order KdV equation In this section, by using the optimal system, we give some group-invariant solutions. 125 3.4.1. One-dimensional optimal system of subalgebras In order to get the optimal system, we applying the adjoint transformations formula [1] given by 129 Ad (exp(Vi ))V j = V j −  [Vi , V j ] + 124 126 127 128 130 131 1 2  [Vi , [Vi , V j ]] − · · · 2 (30) where  is a nonzero constant. Here [Vi , Vj ] is the commutator for the Lie algebra given by [Vi , V j ] = ViV j − V jVi . V1 , V2 + λV1 , V3 . 132 133 (31) We can get an optimal system of one-dimensional subalgebras: 134 135 (32) 3.4.2. Symmetry reductions In the present subsection, we employ the optimal system of one-dimensional subalgebras to deal with (1). and in the next subsection we will give some exact solutions of (1). (1) V1 . For the generator V1 , one can get the group-invariant solution is u = f (ξ ), and ξ = t is the group-invariant, in this case, one can get trivial solution u(x, t ) = C, and C is a constant quantity. (2) V2 + λV1 . For the case of V2 + λV1 , we get the group-invariant solutions u = f ( ξ ), 136 137 138 139 140 (27) −λ f  + α f 2 f  + β f  f  + γ f f  + f (5) = 0. 141 142 143 144 145 146 147 148 (33) where ξ = x − λt. Plugging (33) into (2), one can get the following ODE: Suppose (1) can be written an a conservation law, Dt T (x, t, u ) + Dx X (x, t, u ) = 0. 117 (25) Then, solving the overdetermined equations, one can get τ = 0, ξ = 1, η = 0. 110 The PDE system S(x, t, u, v ) = 0 given by Solving the overdetermined system of equations, leads to η= 101 3 149 150 (34) 116 151 Please cite this article as: G. Wang et al., Group analysis, exact solutions and conservation laws of a generalized fifth order KdV equation, Chaos, Solitons and Fractals (2016), http://dx.doi.org/10.1016/j.chaos.2016.02.013 ARTICLE IN PRESS JID: CHAOS 4 G. Wang et al. / Chaos, Solitons and Fractals xxx (2016) xxx–xxx In particular, if β = 2γ , and each term is multiplied by f, and integral once, one can get 152 153 − λ 2 2 1 4 α f + γ f 2 f  + f f (4) − f  f  4 f + (35) 159 160 1 (36) Substitution of (36) into (2), one can lead to − 2 1 f − ξ f  + α f 2 f  + β f  f  + γ f f  + f (5) = 0. 5 5 cn ξ n . ∞  +α n=1  n  (38) n  α ∞  n  k  1 2 − ncn − cn 5 5 (n + 1 − k )c j ck− j cn+1−kξ u(x, t ) = n ∞  n  ξ n+5 . (42) 165 (k + 1 ) n=1 1 = 2 1 1 1 1 n=0  (39) × α n  k  (n + 1 − k )c j ck− j cn+1−k k=0 j=0 (40) +β n  (k + 1 )(n + 1 − k )(n + 2 − k )ck+1 cn+2−k k=0 +γ 1 (n + 1 )(n + 2 )(n + 3 )(n + 4 )(n + 5 ) α 1 cn+5 (xt − 5 )n+5 (t − 5 ) +c4 (xt − 5 )4 ∞  1 − (n + 1 )(n + 2 )(n + 3 )(n + 4 )(n + 5 ) For general case, if n ≥ 1, one can get n  k  ∞  c0 + c1 (xt − 5 ) + c2 (xt − 5 )2 + c3 (xt − 5 )3 Comparing coefficients for n = 0 in (39), one yields  1 n=0 n=1 c0 − α c02 c1 − 2β c1 c2 − 6γ c0 c3 . 120 1 c0 + c1 (xt − 5 ) + c2 (xt − 5 )2 + c3 (xt − 5 )3 +c4 (xt − 5 )4 + n=1 k=0 ×  1 × (n + 3 − k )ck cn+3−k ξ n ∞ ∞ 2  1 2 − ncn ξ n − c0 + cn ξ n = 0. 5 5 5 cn+5 = − (n + 1 − k )(n + 2 − k )(n + 3 − k )ck cn+3−k Therefore, one can get × (n + 1 − k )(n + 2 − k )ck+1 cn+2−k ξ n ∞  n  +6γ c0 c3 + γ (n + 1 − k )(n + 2 − k ) 2 (k + 1 )(n + 1 − k )(n + 2 − k )ck+1 cn+2−k k=0 n=1 k=0 c5 = 5 (n + 1 − k )c j ck− j cn+1−k (n + 1 )(n + 2 )(n + 3 )(n + 4 )(n + 5 )cn+5 ξ n +2β c1 c2 + β 163 n  k  k=0 j=0 n=1 k=0 j=0 162 α +γ n=1 c02 c1 + 4 k=0 Substituting (38) into (36), one can have 120c5 + 3 c0 − α c02 c1 − 2β c1 c2 − 6γ c0 c3 5 ξ 120 ∞  1 − (n + 1 )(n + 2 )(n + 3 )(n + 4 )(n + 5 ) +β n=0 161 = c0 + c1 ξ + c2 ξ + c3 ξ + c4 ξ (37) Supposing that (34) has the following solutions ∞  n=1 2 × 3.5. Exact group-invariant solutions using power serious method f (ξ ) = ∞  cn+5 ξ n+5 +5 u = t − 5 f (ξ ), ξ = xt − 5 . 157 164 2 where k is an integration constant. 155 (3) V3 . 156 In the case of the generator V3 , we get 154 2 In this way, the power series solution of can be rewritten f ( ξ ) = c0 + c1 ξ + c2 ξ 2 + c3 ξ 3 + c4 ξ 4 + c5 ξ 5 + 1 + ( f  )2 + k = 0, 2 158 [m3Gdc;February 24, 2016;10:55] n  k=0 (n + 1 − k )(n + 2 − k )(n + 3 − k )ck cn+3−k  1 2 1 2 − ncn − cn (xt − 5 )n+5 (t − 5 ), 5 5 (n + 1 − k )c j ck− j cn+1−k (43) k=0 j=0 +β n  here ci (i = 0, 1, 2, 3, 4 ) are arbitrary constants. (k + 1 )(n + 1 − k )(n + 2 − k )ck+1 cn+2−k 166 k=0 +γ n  k=0 (n + 1 − k )(n + 2 − k )(n + 3 − k )ck cn+3−k  1 2 − ncn − cn . 5 5 (41) Remark. The exact solution of (43) also can be fixed in similarly way. The details are omitted here. In addition, by using the Maple soft, one can get following Jacobi function and Weierstrass elliptic function solutions: Please cite this article as: G. Wang et al., Group analysis, exact solutions and conservation laws of a generalized fifth order KdV equation, Chaos, Solitons and Fractals (2016), http://dx.doi.org/10.1016/j.chaos.2016.02.013 167 168 169 170 171 Q4 ARTICLE IN PRESS JID: CHAOS [m3Gdc;February 24, 2016;10:55] G. Wang et al. / Chaos, Solitons and Fractals xxx (2016) xxx–xxx  −6C32 SN −6 u(x, t ) = 2βγ +β 2 −  2 β +4β 3 γ +4β 2 γ 2 −40αβ 2 −12α C3 5 t + C x + C , i 3 2 α √ 4   β 12 γ 2 + 12 β γ + 6 β 2 + 6 β 4 + 4 β 3 γ + 4 β 2 γ 2 − 40 α β 2 − 120 α  −6γ 2 2βγ +β 2 − × ×  √ 4 β +4β 3 γ +4 β 2 γ 2 −40αβ 2 −12α α −60β 2 − 60   −6C32 SN −6 2βγ +β 2 + β 4 + 4β 3 γ + 4β 2 γ 2 − 40αβ 2  2 β +4β 3 γ +4β 2 γ 2 −40αβ 2 −12α C3 5 t + C3 x + C2 , i α × ×  √ 4 β +4β 3 γ +4 β 2 γ 2 −40αβ 2 −12α α −60β 2 + 60 1 u(x, t ) = −3℘  2   β 2 + 4βγ + 4γ 2 − 40α C4 α . −1 2 (46) 173 1 u(x, t ) = −3℘  2 C4 5C2 (−3N + 36 )t + C4 x + C3 , C2 , C1 × 2γ +β + β 2 +4 βγ +4γ 2 −40   −1 181 182 183 184 There are four types of solutions that are going to be extracted for (48) with the aid of the method of undetermined coefficients. They are in the next few subsections: p (49) with τ = B(x − vt ) (50) 189 190 −(1 + p)(2 + p)(3 + p)(4 + p)δ B sech 4 τ τ =0 p+3 p+5 (51) p= 4 . n 191 192 (52) Notice also that the same principle allows to equate 2 p + 3 with p + 5, and 2 p + 1 with p + 3, both situations leading to p=2 193 194 195 (53) Thus, from (52) and (53) we have 196 (54) Consequently, the system (51) collapses into In order to obtain solitary waves, the starting hypothesis is: 188 τ n = 2. 4.1. Solitary waves u(x, t ) = Asech τ 185 (48) 187 τ The balancing principle allows to equate the exponents (n + 1 ) p + 1 with p + 5 from which This section will obtain solitary wave solutions to the model equation given by (1). The method of undetermined coefficients will be adopted to retrieve these solitons. In this case we rewrite Eq. (1) as follows: 179 (n+1 ) p+1 +2(1 + p)(2 + p)[2 + p(2 + p)]δ B sech 175 186 (v − p4 δ B4 )sech p+1 τ − p2 (β + γ )AB2 sech2 p+1 τ 4 4. Solitons solutions 180 (45) 2 p+3 174 ut + α un ux + β ux uxx + γ uuxxx + δ uxxxxx = 0. . +(1 + p)(2γ + p(β + γ ))AB2 sech α C4 α . 2  where A is the amplitude of the soliton, p > 0 is a parameter that will be obtained with the aid of the balancing principle, B represents the inverse width of the soliton while v is the speed. By substituting (49) into (48) one obtain −α An sech (47) 177 − 72γ 2 + 120βγ β 4 + 4β 3 γ + 4β 2 γ 2 − 40αβ 2 C4 5C2 (−3M + 36 )t + C4 x + C3 , C2 , C1 × 2γ + β − (44)  β 12 γ 2 + 12βγ + 6β 2 − 6 β 4 + 4 β 3 γ + 4 β 2 γ 2 − 40 α β 2 − 120 α 172 .  β 12 γ 2 + 12 β γ + 6 β 2 + 6 β 4 + 4 β 3 γ + 4 β 2 γ 2 − 40 α β 2 − 120 α   √ 4 β 12 γ 2 + 12 β γ + 6 β 2 + 6 β 4 + 4 β 3 γ + 4 β 2 γ 2 − 40 α β 2 − 120 α  178 − 72γ 2 + 120βγ β 12 γ 2 + 12 β γ + 6 β 2 + 6 β 4 + 4 β 3 γ + 4 β 2 γ 2 − 40 α β 2 − 120 α −6γ 2 2βγ +β 2 + 176  β 12 γ 2 + 12 β γ + 6 β 2 + 6 β 4 + 4 β 3 γ + 4 β 2 γ 2 − 40 α β 2 − 120 α  u(x, t ) = 5 197 (v − 16δ B )sech τ − 4[(β + γ )A − 60δ B ]B sech τ 4 3 2 2 −[360δ B4 + α A2 − 6(2γ + β )AB2 ]sech τ = 0. 7 5 (55) After equating the coefficients of the linearly independent functions sechj τ for j = 3, 5, 7 to zero one get v = 16δ B , 4 (56) Please cite this article as: G. Wang et al., Group analysis, exact solutions and conservation laws of a generalized fifth order KdV equation, Chaos, Solitons and Fractals (2016), http://dx.doi.org/10.1016/j.chaos.2016.02.013 198 199 ARTICLE IN PRESS JID: CHAOS 6 200 201 G. Wang et al. / Chaos, Solitons and Fractals xxx (2016) xxx–xxx A= 60δ B2 β +γ 205 4.3. Singular solitary waves (Type-I) 229 For type-I singular solitary wave solution to Eq. (48) the starting hypothesis is 230 (59) Therefore, the solitary wave solution to (48) is given by 2 206 (58) and the identity u(x, t ) = Asech [B(x − vt )] 204 226 constrained by (2γ + β )(γ + β ) − (β + γ )2 − 10αδ = 0. 203 where the amplitude is given in (65) while the speed turns to be (64) along with the corresponding conditions. In addition, the condition (66) has to be satisfied. (57) β + γ = 0, 202 (60) where the speed v is given in (56), while the amplitude A is provided in (57). For the solution to exist the identity (59) has to be satisfied along with the condition (58). u(x, t ) = Acsch τ p 207 4.2. Shock waves 208 In order to solve the generalized fifth order KdV Eq. (48) for shock wave the starting hypothesis is taken to be 210 211 212 213 u(x, t ) = A tanh τ p (61) with τ as defined in (50), A and B are free parameters, and p > 0 is a key parameter to be determined. Substituting Eq. (49) into Eq. (48) gives in a simplified form τ − α An tanh(n+1) p−1 τ p−3 +4( p − 1 )( p − 2 )[2 + p( p − 2 )]δ B4 tanh τ p−5 4 −( p − 1 )( p − 2 )( p − 3 )( p − 4 )δ B tanh τ 2 p−1 2 2 +2 p (β + γ )AB tanh τ + ( p − 1) 2 p−3 ×[2γ − p(β + γ )]AB2 tanh τ p+1 +4( p + 1 )( p + 2 )[2 + p(2 + p)]δ B4 tanh τ p+3 4 −( p + 1 )( p + 2 )( p + 3 )( p + 4 )δ B tanh τ 2 p+1 −( p + 1 )[2γ + p(β + γ )]AB2 tanh τ = 0. (62) [v − 2 p2 (5 + 3 p2 )δ B4 ] tanh 214 215 216 217 218 p−1 By the balancing principle, equating the exponents (n + 1 ) p − 1 and p + 3 leads to (52), but also it is possible to equate 2 p + 1 with p + 3 resulting in (53) and consequently one get (54). In view of this values of p and n, the Eq. (62) can be rewritten as [v − 136δ B4 − 2β AB2 ] tanh τ + 8[(β + γ )A + 60δ B2 ]B2 tanh τ − [6(β + 2γ )AB2 3 + 360δ B + α A ] tanh τ = 0. 4 219 220 2 5 (63) Equating to zero the coefficients of the linearly independent functions tanh j τ for j = 1, 3, 5 lead us to v = 2(β A + 68δ B2 )B2 , 221 and the amplitude becomes A=− 222 60δ B2 . β +γ (65) along with the condition (58). In addition, the identity 224 225 (66) has to be satisfied in order for the shock waves to exist. Therefore, the shock wave to the generalized fifth order KdV Eq. (48) is given by u(x, t ) = A tanh [B(x − vt )] 2 (67) 231 232 233 234 235 (v − p δ B )csch τ − p (β + γ )AB csch τ (n+1 ) p+1 n −α A csch τ − 2(1 + p)(2 + p) p+3 ×(2 + p(2 + p))δ B4 csch τ 2 p+3 −(1 + p)[2γ + p(β + γ )]AB2 csch τ p+5 −(1 + p)(2 + p)(3 + p)(4 + p)δ B4 csch τ = 0. (69) p+1 4 2 2 p+1 2 As in Eq. (51), it is possible to equate (n + 1 ) p + 1 with p + 5, 2 p + 3 with p + 5, and also 2 p + 1 with p + 3, thus leading to (53), and consequently (54). As a consequence, the Eq. (69) becomes 236 237 238 239 (v − 16δ B )csch τ − 4[(β + γ )A + 60δ B ]B csch τ 7 −[360δ B4 + α A2 + 6(2γ + β )AB2 ]csch τ = 0. (70) 3 4 2 2 5 After equating the coefficients of the linearly independent functions cschj τ for j = 3, 5, 7 to zero one get the wave speed as in (56), the amplitude as in (65), and the identity turns to be the same as in (66). Thus, the type-I singular solitary wave solution to the generalized KdV Eq. (48) is given by u(x, t ) = Acsch [B(x − vt )] 2 240 241 242 243 244 245 (71) where the soliton speed and amplitude are given in (56) and (65) respectively, while the identity (66) has to be satisfied in order to preserve the existence of the solitary wave. 246 247 248 249 4.4. Singular solitary waves (Type-II) 250 To obtain type-II singular solitary wave solutions to Eq. (48) the starting Ansatz is 252 u(x, t ) = A coth τ p [v − 2 p (5 + 3 p )δ B ] coth (n+1 ) p−1 τ − α A coth p−3 +4( p − 1 )( p − 2 )[2 + p( p − 2 )]δ B4 coth τ p−5 −( p − 1 )( p − 2 )( p − 3 )( p − 4 )δ B4 coth τ 2 p−1 +2 p2 (β + γ )AB2 coth τ + ( p − 1) 2 p−3 2 ×[2γ − p(β + γ )]AB coth τ p+1 +4( p + 1 )( p + 2 )[2 + p(2 + p)]δ B4 coth τ p+3 4 −( p + 1 )( p + 2 )( p + 3 )( p + 4 )δ B coth τ 2 p+1 −( p + 1 )[2γ + p(β + γ )]AB2 coth τ = 0. 2 251 (72) where the meaning of A, p and τ are as in the previous subsections. By inserting (72) into the fifth order KdV Eq. (48) one get 2 (β + γ )2 − (2γ + β )(γ + β ) + 10αδ = 0. 223 (64) 227 228 (68) where the parameter p > 0 will be determined with the help of the balancing principle, A and B are free parameters, while τ retain the same meaning as in (50). The substitution of (68) into (48) leads to 4 209 [m3Gdc;February 24, 2016;10:55] 4 p−1 n τ (73) Please cite this article as: G. Wang et al., Group analysis, exact solutions and conservation laws of a generalized fifth order KdV equation, Chaos, Solitons and Fractals (2016), http://dx.doi.org/10.1016/j.chaos.2016.02.013 253 254 255 ARTICLE IN PRESS JID: CHAOS [m3Gdc;February 24, 2016;10:55] G. Wang et al. / Chaos, Solitons and Fractals xxx (2016) xxx–xxx 256 257 The balancing principle leads to (53) and (54). Both values reduce the Eq. (73) into [v − 136δ B4 − 2β AB2 ] coth τ + 8[(β + γ )A + 60δ B2 ]B2 coth τ − [6(β + 2γ )AB2 follows:  ∞ 5 Q5 258 259 260 and consequently, the results (64)–(66) reappear. Finally, the type-II singular solitary wave solution to the generalized fifth order KdV Eq. (48) is given by u(x, t ) = A coth [B(x − vt )] 2 261 262  ∞ (74) (75) where the amplitude is provided in (65) while the speed turns to be (64) along with the corresponding conditions. 263 5. Conservation laws 264 In this section, we employ the multipliers method [3– 5] to deal with the conservation law. Firstly, we introduces some basic definitions and concepts. From what has been described above, suppose the conservation law is given by Dx T x + Dt T t = on the solutions of (1). In Section 2, we got the Lie point symmetry generators X1 = ∂t and X2 = ∂x . For the case n = 2, we obtain the additional scaling symmetry X3 = x∂x + 5t ∂t − 2u∂u . In general, we have the only conserved vector based on the multiplier Q = 1 given by 265 266 267 268 269 270 271 272 273 274 1 1 α un+1 + (β − γ )ux 2 + γ uuxx + uxxxx , n+1 2 T t = u. Tx = 275 276 (76) In particular, if β = 2γ , one can get the multiplier Q = u, and get the conserved vector 1 1 α un+2 + γ u2 uxx + uuxxxx − ux uxxx + u2xx , n+2 2 1 T t = u2 . (77) 2 Tx = 277 For n = 2, subject to the condition 10α + 2β 2 − 7βγ + 3γ 2 = 0, 278 279 280 1 (α (8β − 4γ )u5 + 20(2β − γ )γ u3 uxx 200 −20u(5uxt − 2((β + 2γ )uxx 2 + (−2β + γ )ux uxxx )) −5u2 ((10α + 2β 2 − 7βγ + 3γ 2 )ux 2 +4(−2β + γ )uxxxx ) + 20(5ut ux + 2(2β − γ )ux 2 uxx −5(uxxx 2 − 2uxx uxxxx ))), 1 Tt = ((2β − γ )u3 + 15uuxx ). 30 281 282 283 284 285 1 I2 = T t dx = 2 −∞  ∞ 287  ∞ 1 I3 = T t dx = 30 −∞ = (80) 2A2 u2 d x = 3B −∞  ∞ −∞ (81) 288  (2β − γ )u3 + 15uuxx dx 8A2 {(2β − γ )A + 15B2 } 225B (82) Remark. It does not exist other nth order multiplier, in other words, it only exist, for the general case, second order multiplier. 289 6. Conclusion 292 In the present paper, using the group methods, and the multiplier approach, the generalized fifth-order KdV equation is studied. Furthermore, we derive the corresponding Lie algebra and the similarity reductions of special case of generalized fifth-order KdV equation. Also, we found that the analyzed model does not admit nonclassical type symmetries for n = 2. In addition, on the basis of symmetries, the optimal system is constructed, based on the optimal system, some exact solutions are presented. Meanwhile, some soliton solutions are presented. Finally, conservations laws are derived. These results are important for the understanding of nonlinear interaction and behaviors of complex system in some piratical physical problems. Acknowledgments This work is supported by the Research Project International Graduate Exchange Program of Beijing Institute of Technology, National Natural Science Foundation of China (NNSFC) (Grant no. 11171022), Graduate Student Science and Technology Innovation Activities of Beijing Institute of Technology (No. 2014cx10037). 290 291 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 (78) we obtain an additional, second order, multiplier Q = 1 2 10 ((2β − γ )u + 10uxx ) leading to the nontrivial conserved flow Tx = 286  ∞ 2A I1 = T t dx = udx = B −∞ −∞ 3 + 360δ B4 + α A2 ] coth τ = 0. 7 (79) We note that the action of X3 on Q satisfies X3 Q = −4Q. For any solution u(x, t), if u and its derivatives converge to ∞ x −→ ±∞, −∞ T t dx provide conserved quantities. From the solitary wave solution derived in the previous section, the conserved quantities, with δ = 1, are as References 313 [1] Olver PJ. Application of Lie group to differential equation. New York: Springer; 1986. [2] Ovsiannikov LV. Group analysis of differential equations. New York: Academic Press; 1982. [3] Bluman GW, Kumei S. Symmetries and differential equations. New York: Springer; 1989. [4] Bluman GW, Cheviakov A, Anco S. Applications of symmetry methods to partial differential equations. New York: Springer; 2010. [5] Bluman GW, Anco S. Symmetry and integration methods for differential equations. New York: Springer; 2002. [6] Ibragimov N.H., editor. CRC handbook of Lie group analysis of differential equations; vol. 1–3. Boca Raton, FL: CRC Press; 1994. [7] Wazwaz AM. Soliton solutions for the fifth-order KdV equation and the Kawahara equation with time-dependent coefficients. Phys Scr 2010;82:035009. [8] Liu HZ. Complete group classifications and symmetry reductions of the fractional fifth-order KdV types of equations. Stud Appl Math 2013;131:317–30. [9] Chaolu T, Bluman GW. An algorithmic method for showing existence of nontrivial nonclassical symmetries of partial differential equations without solving determining equations. J Math Anal Appl 2014;411:281–96. 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 Please cite this article as: G. Wang et al., Group analysis, exact solutions and conservation laws of a generalized fifth order KdV equation, Chaos, Solitons and Fractals (2016), http://dx.doi.org/10.1016/j.chaos.2016.02.013 JID: CHAOS 8 336 337 338 Q6 339 340 341 342 343 344 345 346 347 ARTICLE IN PRESS [m3Gdc;February 24, 2016;10:55] G. Wang et al. / Chaos, Solitons and Fractals xxx (2016) xxx–xxx [10] Bluman GW, Chaolu T. Conservation laws for nonlinear telegraph equations. J Math Anal Appl 2005;310:459–76. [11] Anco SC, Bluman GW. Direct construction of conservation laws from field equations. Phys Rev Lett 1997;78:2869–73. [12] Hirota R.. The direct method in soliton theoryCambridge:Cambridge University Press; 2004. [13] Wang GW. Symmetry analysis and rogue wave solutions for the (2+1)-dimensional nonlinear Schrödinger equation with variable coefficients. Appl Math Lett 2016;56:56–64. [14] Wang GW, Kara AH. Nonlocal symmetry analysis, explicit solutions and conservation laws for the fourth-order burgers’ equation. Chaos Solitons Fractals 2015;81:290–8. [15] Razborova P, Kara AH, Biswas A. Additional conservation laws for Rosenau-KdV-RLW equation with power law nonlinearity by lie symmetry. Nonlinear Dyn 2015;79:743–8. [16] Wang GW, Fakhar K. Lie symmetry analysis, nonlinear selfadjointness and conservation laws to an extended (2+1)dimensional Zakharov–Kuznetsov–Burgers equation. Comput. Fluids. 2015;119:143–8. [17] Kara AH. On the reduction of some dispersionless integrable systems. Acta Appl Math 2014;132:371–6. Please cite this article as: G. Wang et al., Group analysis, exact solutions and conservation laws of a generalized fifth order KdV equation, Chaos, Solitons and Fractals (2016), http://dx.doi.org/10.1016/j.chaos.2016.02.013 348 349 350 351 352 353 354 355 356