数学物理学报  2018, Vol. 38 Issue (4): 658-670   PDF    
扩展功能
加入收藏夹
复制引文信息
加入引用管理器
Email Alert
RSS
本文作者相关文章
白喜瑞
广义耦合方程的延拓结构
白喜瑞     
宁波大学非线性研究中心 浙江 宁波 315211
摘要:该文利用延拓结构理论及单(半单)Lie代数的性质,研究了两组对偶系统的延拓结构.并且利用Lie代数表示理论,给出了两组对偶系统的Lax对表示.对于Camassa-Holm(CH)型的方程,通过对其超定方程的分析,仅仅选择了阶数小于等于2的函数F进行讨论,然而经过计算与分析却只存在阶数为1的情况.
关键词广义耦合方程    延拓结构    对偶系统    Lax对    
Prolongation Structures of Generalized Coupled Equation
Bai Xirui     
Center for Nonlinear Studies, Ningbo University, Zhejiang Ningbo 315211
Abstract: In this paper, we adopt the theory of continuation structure and properties of simple or semi-simple Lie algebra and studied the prolongation structure for two dual systems. Furthermore, the Lax pair of these equations are obtained by the method of Lie representation theory. Considering the CH type equation, based on analysing its determined equations, we select the functions F that ord(F) ≤ 2. However, there merely exists one order situation through calculation and analysis.
Key words: Generalized coupled equation     Prolongation structure     Dual system     Lax pair    
1 引言

在过去的几十年里, 许多非线性发展方程被用于研究复杂的物理现象.这些方程拥有双哈密顿结构, 递归算子和无穷多广义对称.迄今为止, 人们发现了许多方法来检验方程的可积性, 例如双哈密顿结构, 无穷守恒律, Lax对等.本文中, 如果一个方程拥有Lax对, 那么我们就说它是可积的.

1996年, Olver和Rosenau利用重组哈密顿算子的方法构造了对偶系统.在文献[1]中, 它们给出了一些具体的例子, 例如KdV方程的对偶系统为CH方程, mKdV方程的对偶系统为mCH方程等. 1975年, Wahlquist和Estabrook[2]第一次利用延拓结构方法得到了KdV的延拓结构为

$ F=2X_1+2uX_2+3u^2X_3, $
$ G=-2\left(\hat{p}+6u^2\right)X_2+3\left(z^2-8u^3-2u\hat{p}\right)X_3+8X_4+8uX_5+4u^2X_6+4zX_7, $

这里$z=u_x, \; \hat{p}=u_{xx}$, $X_i, \; i=1, \cdots, 7$是相应的延拓代数.延拓结构方法是构造非线性发展方程线性谱问题的一个有效的工具, 因此吸引了许多学者的研究.文献[3]给出了非线性薛定谔方程的延拓结构.文献[4]给出了势KdV方程的延拓结构.更多方程的延拓结构可参见文献[5-8].在求CH方程的延拓结构时, 由于在最终所得的超定方程中, 未知函数$F$$G$包含在同一个方程中, 因此我们不能够直接解出函数$F$$G$的具体表达式.但是, 当我们取定其中一个未知函数时, 另一个函数自然就确定下来.对于这方面的研究, 可参见文献[9-10].在文献[11]中, 作者把未知函数$G$严格限制为变量$v=u_x$的线性函数, 得到了KdV方程的对偶系统CH方程的延拓结构为

$ F=X_1+uX_2-wX_2+uwX_3-\frac{1}{2}u^2X_3-\frac{1}{2}w^2X_3, $
$ G=\left(uw-u^2\right)X_2+\left(\frac{1}{2}uw^2-u^2w+\frac{1}{2}u^3\right)X_3+vX_4+X_5+uX_6, $

其中$v=u_x, \; w=u_{xx}$.延拓结构理论更是广泛的应用于耦合KdV方程[12-18], 由对偶系统理论可知, 研究耦合CH型方程的延拓结构也是一个有意义的课题.

本文在第二节主要研究了广义耦合KdV方程(2.1)的延拓结构并通过将延拓代数嵌入到单Lie代数${\rm sl}(4, {\Bbb C})$中得到了此方程的线性谱问题.在第三节中根据文献[11]中的方法研究了方程(2.1)的对偶系统的延拓结构及Lax对.在第四节中给出了另外一个CH型方程的延拓结构及Lax对.

2 广义耦合KdV方程的延拓结构

在开始求延拓结构之前, 我们先介绍一个Lie代数中的基本定理, 这在后面求解Lax对中起着重要的作用.

定理2.1 设$ X$$Y$是李代数${\mathfrak g}={\rm sl}(n+1, {\Bbb C})$中的两个元素, 如果$X, Y$满足$[X, Y]=aY, \, (a\neq0)$$X \in {\rm ad} \, Y$.则

$ Y =e_{\pm}, \;\;\;\; X = \pm \frac{1}{2}ah, $

其中$e_{\pm}$${\mathfrak g}$的幂零元素, $h $${\mathfrak g}$的零元素[19-20].

在本节中, 我们主要研究下面广义耦合KdV方程的延拓结构

$ \begin{equation}\label{e2.1}\begin{array}{ll} u_t+u_{xxx}+3uu_x+uv_x+u_xv-vv_x=0, \\ v_t+v_{xxx}+3vv_x+uv_x+u_xv-uu_x=0. \end{array}\end{equation} $ (2.1)

方程(2.1)是广义耦合KdV型方程

$ \begin{eqnarray*} && u_t=\widetilde{F}[u, v], \\ &&v_t=\widetilde{F}[v, u], \end{eqnarray*} $

中的一类, 这里$\widetilde{F}$依赖于$u, \, v$$u, \, v$的各阶导数且权重为3或5.文献[21]指出此方程不能约化为KdV方程$u_t=u_{xxx}+uu_x$并给出了它的双哈密顿结构.

为了把方程(2.1)表达成外微分的形式, 我们先引进一些新的独立变量

$ \begin{equation}\label{e2.2} p=u_x, \quad\quad q=v_x, \quad\quad r=p_x=u_{xx}, \quad\quad s=q_x=v_{xx}. \end{equation} $ (2.2)

利用新的变量(2.2), 我们就可以把方程(2.1)表达成下述二阶外微分的形式

$ \begin{equation}\label{e2.3}\begin{array}{ll} \alpha_1&={\rm d}u\wedge {\rm d}t-p{\rm d}x\wedge {\rm d}t, \quad \alpha_2={\rm d}v\wedge {\rm d}t-q{\rm d}x\wedge {\rm d}t, \\ \alpha_3&={\rm d}p\wedge {\rm d}t-r{\rm d}x\wedge {\rm d}t, \quad \alpha_4={\rm d}q\wedge {\rm d}t-s{\rm d}x\wedge {\rm d}t, \\ \alpha_5&={\rm d}u\wedge {\rm d}x-{\rm d}r\wedge {\rm d}t+(3up+uq+vp-vq){\rm d}t\wedge {\rm d}x, \\ \alpha_6&={\rm d}v\wedge {\rm d}x-{\rm d}s\wedge {\rm d}t+(3vq+uq+vp-up){\rm d}t\wedge {\rm d}x, \end{array}\end{equation} $ (2.3)

这里${\rm d}$表示外微分, $\wedge$表示外积(反对称张量积)[22].易证理想$I=\{\alpha_1, \alpha_2, \alpha_3, \alpha_4, \alpha_5, \alpha_6\}$是一个闭理想.

$ \begin{equation}\label{e2.4}\begin{array}{ll} {\rm d}\alpha_1&={\rm d}x\wedge \alpha_3, \qquad {\rm d}\alpha_2={\rm d}x\wedge \alpha_4, \\ {\rm d}\alpha_3&=-{\rm d}x\wedge \alpha_5, \qquad {\rm d}\alpha_4=-{\rm d}x\wedge \alpha_6, \\ {\rm d}\alpha_5&=(3p+q){\rm d}x\wedge \alpha_1+(p-q){\rm d}x\wedge \alpha_2+(3u+v){\rm d}x\wedge \alpha_3+(u+v){\rm d}x\wedge \alpha_4, \\ {\rm d}\alpha_6&=-(p-q){\rm d}x\wedge \alpha_1+(p+3q){\rm d}x\wedge \alpha_2-(u-v){\rm d}x\wedge \alpha_3+(u+3v){\rm d}x\wedge \alpha_4. \end{array}\end{equation} $ (2.4)

$ \begin{equation}\label{e2.5} \omega_k={\rm d}y^k-F^k(u, v, p, q, r, s;y^k){\rm d}x-G^k(u, v, p, q, r, s;y^k){\rm d}t, \end{equation} $ (2.5)

其中$y^k \, (k=1, 2, \cdots, n)$被称为伪势.假定$F^k$$G^k$具有下面的形式

$ \begin{equation}\label{e2.6} F^k=F^k_jy^j, \qquad G^k=G^k_jy^j, \end{equation} $ (2.6)

在后面的讨论中, 我们把$F^k_j$$G^k_j$分别简记为$F$$G$.

为了得到方程(2.1)的延拓结构, $I\cup{\omega_k}$也必须是一个闭理想, 即

$ \begin{equation}\label{e2.7} {\rm d}\omega_k=\sum\limits_{i=1}^6 f^k_i\alpha_i+\sum\limits_j\left(A^k_j{\rm d}x+B^k_j{\rm d}t\right)\wedge \omega_j, \end{equation} $ (2.7)

这里$f^k_i, \, A^k_j, \, B^k_j$均为待定函数.

结合方程(2.5)与方程(2.7), 我们可得到如下一系列非线性偏微分方程

$ \begin{equation}\label{e2.8}\begin{array}{ll} F_p=F_q=F_r=F_s=0, \quad\quad F_u+G_r=0, \quad\quad F_v+G_s=0, \\ pG_u+qG_v+rG_p+sG_q+(3up+uq+vp-vq)F_u+(3vq+uq+vp-up)F_v-[F, G]=0, \end{array}\end{equation} $ (2.8)

其中$[F, G]=FG-GF$.

经过对(2.8)式中方程的分析与计算, 我们可解得

$ \begin{equation}\label{e2.9}\begin{array}{ll} F&=X_1+uX_2+vX_3, \\ G&=-rX_2-sX_3-pX_4-qX_5+X_6+u^2X_7+v^2X_8+uX_9+vX_{10}+uvX_{11}, \end{array}\end{equation} $ (2.9)

并满足可积条件

$ \begin{equation}\label{e2.10}\begin{array}{ll} &[X_1, X_6]=[X_2, X_3]=[X_2, X_7]=[X_3, X_8]=0, \\ &[X_1, X_2]=X_4, \quad \quad [X_1, X_3]=X_5, \quad\quad [X_4, X_1]=X_9, \quad\quad [X_5, X_1]=X_{10}, \\ &[X_2, X_5]=[X_3, X_4], \quad\quad [X_1, X_7]+[X_2, X_9]=0, \quad\quad [X_1, X_8]+[X_3, X_{10}]=0, \\ &[X_1, X_9]+[X_2, X_6]=0, \quad [X_1, X_{10}]+[X_3, X_6]=0, \quad [X_2, X_8]+[X_3, X_{11}]=0, \\ &[X_2, X_{11}]+[X_3, X_7]=0, \quad \quad [X_1, X_{11}]+[X_2, X_{10}]+[X_3, X_9]=0, \\ &2X_7+3X_2-X_3+[X_2, X_4]=0, \quad \quad 2X_8-X_2+3X_3+[X_3, X_5]=0, \\ &X_2+X_3+[X_3, X_4]+X_{11}=0, \quad \quad X_2+X_3+[X_2, X_5]+X_{11}=0, \end{array}\end{equation} $ (2.10)

这里所有的矩阵$X_i \, (i=1, 2, \cdots, 11)$都为待定矩阵.它们决定了一个不完全的李代数$L$, 我们把这个李代数称为延拓代数.

通过Jacobi恒等式, 我们可以进一步的得到$X_i \, (i=1, 2, \cdots, 11)$之间的关系

$ \begin{equation}\label{e2.11}\begin{array}{ll} &2[X_1, [X_3, X_4]]=[X_1, X_{11}], \quad\quad [X_3, X_7]=0, \\ &[X_2, X_8]=0, \quad\quad [X_2, X_{11}]=[X_3, X_{11}]=0. \end{array}\end{equation} $ (2.11)

因此, 由(2.10)与(2.11)式我们可解得

$ \begin{equation}\label{e2.12} X_{11}=-\frac{2}{3}(X_2+X_3), \quad X_7=-X_2+\frac{1}{3}X_3, \quad X_8=\frac{1}{3}X_2-X_3.\\ \end{equation} $ (2.12)

为了得到与之相关的谱问题, 我们需引入谱参数$\lambda$.方程(2.1)有下面的尺度对称

$ x \rightarrow \lambda^{-1}x, \quad\quad t \rightarrow \lambda^{-3}t, \quad\quad u \rightarrow \lambda^{2}u, \quad\quad v \rightarrow \lambda^{2}v, $

为了使$\omega_k$在这个变换下也是不变的, 那么由(2.9)式可知$X_i \, (i=1, 2, \cdots, 11)$也必须满足

$ X_1 \rightarrow \lambda X_1, \quad\quad X_2 \rightarrow \lambda^{-1}X_2, \quad\quad X_3 \rightarrow \lambda^{-1} X_3, \quad\quad X_4 \rightarrow X_4, $
$ X_5 \rightarrow X_5, \quad\quad X_6 \rightarrow \lambda^{3}X_6, \quad\quad X_7 \rightarrow \lambda^{-1}X_7, \quad\quad X_8 \rightarrow \lambda^{-1}X_8, $
$ X_9\rightarrow \lambda X_9, \quad\quad X_{10} \rightarrow \lambda X_{10}, \quad\quad X_{11} \rightarrow \lambda^{-1} X_{11}. $

利用Dodd和Fordy[23]的方法, 并结合定理2.1, 通过Maple软件的计算, 我们可得到$X_i \, (i=1, 2, \cdots, 11)$的矩阵表示为

$ X_1=\left(\begin{array}{ccccc} 0&~ -\frac{1}{3}~&0&0\\ \lambda &0&0&0\\ 0&\frac{1}{6}&0&-\frac{1}{3}\\ 0&0&\lambda &0 \end{array}\right), \quad\quad X_2=\left(\begin{array}{ccccc} 0&~0~&0&~0\\ 1&0&0&~0\\ 0&0&0&~0\\ 0&0&1&~0 \end{array}\right), \quad\quad X_3=\left(\begin{array}{ccccc} 0&~0~&0&~0\\ 1&0&0&~0\\ 0&0&0&~0\\ 1&0&1&~0 \end{array}\right), $
$ X_4=\left(\begin{array}{ccccc} -\frac{1}{3}&~0~&0&~0\\ 0&\frac{1}{3}&0&~0\\ \frac{1}{6}&0&-\frac{1}{3}&~0\\ 0&-\frac{1}{6}&0&~\frac{1}{3} \end{array}\right), \;X_5=\left(\begin{array}{ccccc} -\frac{1}{3}&~0~&0&~0\\ 0&\frac{1}{3}&0&~0\\ -\frac{1}{6}&0&-\frac{1}{3}&~0\\ 0&\frac{1}{6}&0&~\frac{1}{3} \end{array}\right), \; X_6=\left(\begin{array}{ccccc} 0&-\frac{4}{9}\lambda&0&0\\ \frac{4}{3}\lambda^2&0&0&0\\ 0&\frac{4}{9}\lambda&0&-\frac{4}{9}\lambda\\ -\frac{2}{3}\lambda^2&0&\frac{4}{3}\lambda^2&0 \end{array}\right), $
$ X_7=\left(\begin{array}{ccccc} 0&~0~&0&~0\\ -\frac{2}{3}&0&0&~0\\ 0&0&0&~0\\ \frac{1}{3}&0&-\frac{2}{3}&~0 \end{array}\right), \quad X_8=\left(\begin{array}{ccccc} 0&~0~&0&~0\\ -\frac{2}{3}&0&0&~0\\ 0&0&0&~0\\ -1&0&-\frac{2}{3}&~0 \end{array}\right), \quad X_9=\left(\begin{array}{ccccc} 0&\frac{2}{9}&0&~0\\ \frac{2}{3}\lambda&0&0&~0\\ 0&-\frac{2}{9}&0&~\frac{2}{9}\\ -\frac{1}{3}\lambda&0&\frac{2}{3}\lambda&~0 \end{array}\right), $
$ X_{10}=\left(\begin{array}{ccccc} 0&~\frac{2}{9}~&0&~0\\ \frac{2}{3}\lambda&0&0&~0\\ 0&0&0&~\frac{2}{9}\\ \frac{1}{3}\lambda&0&\frac{2}{3}\lambda&~0 \end{array}\right), \quad X_{11}=\left(\begin{array}{ccccc} 0&~0~&0&~0\\ -\frac{4}{3}&0&0&~0\\ 0&0&0&~0\\ -\frac{2}{3}&0&-\frac{4}{3}&~0 \end{array}\right). $

所以, Lax对的矩阵表示形式为

$ y_x=Fy, \quad y_t=Gy, \quad y=(y^1, y^2, y^3, y^4)^{\rm T}. $

因此, 我们可得到线性谱问题

$ \begin{equation}\label{e2.13}\begin{array}{ll} \left(\begin{array}{ccccc}y^1 \\ y^2 \\ y^3 \\ y^4 \end{array}\right)_x =&\left(\begin{array}{ccccc} 0&~ -\frac{1}{3}~&0&~0\\ u+v+\lambda &0&0&~0\\ 0&\frac{1}{6}&0&~-\frac{1}{3} \\ v&0&u+v+\lambda&~0 \end{array}\right) \left(\begin{array}{ccccc}y^1 \\ y^2 \\ y^3 \\ y^4 \end{array}\right), \\\\ \left(\begin{array}{ccccc}y^1 \\ y^2 \\ y^3 \\ y^4 \end{array}\right)_t =&\left(\begin{array}{ccccc} \frac{1}{3}p+\frac{1}{3}q&~ \frac{2}{9}u+\frac{2}{9}v-\frac{4}{9}\lambda~&0&~0\\ A&-\frac{1}{3}p-\frac{1}{3}q&0&~0 \\ -\frac{1}{6}p+\frac{1}{6}q&-\frac{2}{9}u+\frac{4}{9}\lambda& \frac{1}{3}p+\frac{1}{3}q&~ \frac{2}{9}u+\frac{2}{9}v-\frac{4}{9}\lambda\\ B&\frac{1}{6}p-\frac{1}{6}q&A&~-\frac{1}{3}p-\frac{1}{3}q \end{array}\right) \left(\begin{array}{ccccc}y^1 \\ y^2 \\ y^3 \\ y^4 \end{array}\right), \end{array}\end{equation} $

这里$\lambda$为谱参数,

$ A=-r-s-\frac{2}{3}(u+v)^2+\frac{2}{3}(u+v)\lambda+\frac{4}{3}\lambda^2, $
$ B=-s+\frac{1}{3}u^2-v^2-\frac{2}{3}uv-\frac{1}{3}(u-v)\lambda-\frac{2}{3}\lambda^2. $

方程(2.13)的相容性条件为

$F_t-G_x+[F, G]=0, $

即耦合KdV方程(2.1).

3 第一个广义耦合CH型方程的延拓结构

在本节中, 我们主要研究下列广义耦合CH型方程的延拓结构

$ \begin{equation}\label{e3.1}\begin{array}{ll} m_t=2\left(m+\frac{1}{3}n\right)u_x+m_xu+\frac{1}{3}n_xu+\frac{2}{3}(m-n)v_x+\frac{1}{3}(m_x-n_x)v, \\ n_t=2\left(n+\frac{1}{3}m\right)v_x+n_xv+\frac{1}{3}m_xv+\frac{2}{3}(n-m)u_x+\frac{1}{3}(n_x-m_x)u, \\ m=u-u_{xx}, \quad n=v-v_{xx}. \end{array}\end{equation} $ (3.1)

根据文献[24], 我们可知方程(3.1)为方程(2.1)的对偶系统.为了把方程(3.1)表达成外微分的形式, 我们定义一系列新的独立变量

$ \begin{equation}\label{e3.2} p=u_x, \quad\quad q=v_x, \quad\quad r=u_{xx}, \quad\quad s=v_{xx}. \end{equation} $ (3.2)

则方程(3.1)可被表达成下述二阶外微分的形式

$ \begin{equation}\label{e3.3}\begin{array}{ll} \alpha_1=&{\rm d}u\wedge {\rm d}t-p{\rm d}x\wedge {\rm d}t, \quad \alpha_2={\rm d}v\wedge {\rm d}t-q{\rm d}x\wedge {\rm d}t, \\ \alpha_3=&{\rm d}p\wedge {\rm d}t-r{\rm d}x\wedge {\rm d}t, \quad \alpha_4={\rm d}q\wedge {\rm d}t-s{\rm d}x\wedge {\rm d}t, \\ \alpha_5=& {\rm d}u\wedge {\rm d}x-{\rm d}r\wedge {\rm d}x+\left[\left(3u+v-2r-\frac{2}{3}s\right)p+\left(u-v-\frac{2}{3}r+\frac{2}{3}s\right)q\right]{\rm d}x\wedge {\rm d}t\\ &-\frac{1}{3}(3u+v){\rm d}r\wedge {\rm d}t-\frac{1}{3}(u-v){\rm d}s\wedge {\rm d}t, \\ \alpha_6=& {\rm d}v\wedge {\rm d}x-{\rm d}s\wedge {\rm d}x+\left[-\left(u-v-\frac{2}{3}r+\frac{2}{3}s\right)p+\left(u+3v-\frac{2}{3}r-2s\right)q\right]{\rm d}x\wedge {\rm d}t\\ &+\frac{1}{3}(u-v){\rm d}r\wedge {\rm d}t-\frac{1}{3}(u+3v){\rm d}s\wedge {\rm d}t. \end{array}\end{equation} $ (3.3)

易证理想$I=\{\alpha_1, \alpha_2, \alpha_3, \alpha_4, \alpha_5, \alpha_6\}$是一个闭理想.

$ \begin{equation}\label{e3.4}\begin{array}{ll} {\rm d}\alpha_1=&{\rm d}x\wedge \alpha_3, \qquad {\rm d}\alpha_2={\rm d}x\wedge \alpha_4, \\ {\rm d}\alpha_3=&{\rm d}x\wedge \alpha_1+{\rm d}t\wedge \alpha_5, \qquad {\rm d}\alpha_4={\rm d}x\wedge \alpha_2+{\rm d}t\wedge \alpha_6, \\ {\rm d}\alpha_5=& -\frac{2}{3}(3p+q){\rm d}x\wedge \alpha_1-\frac{2}{3}(p-q){\rm d}x\wedge \alpha_2-\left(3u+v-2r-\frac{2}{3}s\right){\rm d}x\wedge \alpha_3\\ & -\left(u-v-\frac{2}{3}r+\frac{2}{3}s\right){\rm d}x\wedge \alpha_4+\frac{1}{3}(3p+q){\rm d}t\wedge \alpha_5+\frac{1}{3}(p-q){\rm d}t\wedge \alpha_6\\ & +{\rm d}r\wedge \alpha_1+\frac{1}{3}{\rm d}r\wedge \alpha_2+\frac{1}{3}{\rm d}s\wedge \alpha_1-\frac{1}{3}{\rm d}s\wedge \alpha_2, \\ {\rm d}\alpha_6=& \frac{2}{3}(p-q){\rm d}x\wedge \alpha_1-\frac{2}{3}(p+3q){\rm d}x\wedge \alpha_2+\left(u-v-\frac{2}{3}r+\frac{2}{3}s\right){\rm d}x\wedge \alpha_3\\ & -\left(u+3v-\frac{2}{3}r-2s\right){\rm d}x\wedge \alpha_4-\frac{1}{3}(p-q){\rm d}t\wedge \alpha_5+\frac{1}{3}(p+3q){\rm d}t\wedge \alpha_6\\ & -\frac{1}{3}{\rm d}r\wedge \alpha_1+\frac{1}{3}{\rm d}r\wedge \alpha_2+\frac{1}{3}{\rm d}s\wedge \alpha_1+{\rm d}s\wedge \alpha_2. \end{array}\end{equation} $ (3.4)

设一阶外微分

$ \begin{equation}\label{e3.5} \omega_k={\rm d}y^k-F^k(u, v, p, q, r, s;y^k){\rm d}x-G^k(u, v, p, q, r, s;y^k){\rm d}t, \end{equation} $ (3.5)

其中$y^k \, (k=1, 2, \cdots, n)$是伪势.假定$F^k$$G^k$具有下面的形式

$ \begin{equation}\label{e3.6} F^k=F^k_jy^j, \qquad G^k=G^k_jy^j, \end{equation} $ (3.6)

在后面的讨论中, 我们把$F^k_j$$G^k_j$分别简记为$F$$G$.

为了得到方程(3.1)的延拓结构, $I\cup{\omega_k}$也必须是一个闭理想, 即

$ \begin{equation}\label{e3.7} {\rm d}\omega_k=\sum\limits_{i=1}^6 f^k_i\alpha_i+\sum\limits_j(A^k_j{\rm d}x+B^k_j{\rm d}t)\wedge \omega_j, \end{equation} $ (3.7)

这里$f^k_i, \, A^k_j, \, B^k_j$均为待定函数.

则方程(3.1)可化为如下一系列超定方程

$ \begin{equation}\label{e3.8}\begin{array}{ll} F_p=F_q=0, \quad\quad F_u+F_r=0, \quad\quad F_v+F_s=0, \\ G_r=-\frac{1}{3}(3u+v)F_u+\frac{1}{3}(u-v)F_v, \quad\quad G_s=-\frac{1}{3}(u-v)F_u-\frac{1}{3}(u+3v)F_v, \\ pG_u+qG_v+rG_p+sG_q-\left[ \left(3u+v-2r-\frac{2}{3}s\right)p+\left(u-v-\frac{2}{3}r+\frac{2}{3}s\right)q \right] F_u\\ +\left[ \left(u-v-\frac{2}{3}r+\frac{2}{3}s\right)p-\left(u+3v-\frac{2}{3}r-2s\right)q \right] F_v+[F, G]=0, \end{array}\end{equation} $ (3.8)

这里$[F, G]=FG-GF$.

通过分析(3.8)式, 如果我们设

$ \begin{equation}\label{e3.1.1} F=X_1+uX_2+vX_3-rX_2-sX_3, \end{equation} $ (3.9)

那么

$ \begin{equation}\label{e3.1.2} G_r=-\frac{1}{3}(3u+v)X_2+\frac{1}{3}(u-v)X_3, \quad G_s=-\frac{1}{3}(u-v)X_2-\frac{1}{3}(u+3v)X_3. \end{equation} $ (3.10)

根据文献[11]中的方法, 我们可设

$ \begin{equation}\label{e3.1.3} G_p=X_4, \quad G_q=X_5, \end{equation} $ (3.11)

把方程(3.9)和(3.11)代入(3.8)中的第六式, 将所得方程分别关于$p, \, q$微分一次, 我们可解得

$ \begin{eqnarray*} G_u&=&\left(3u+v-2r-\frac{2}{3}s\right)X_2-\left(u-v-\frac{2}{3}r+\frac{2}{3}s\right)X_3+[X_1, X_4]\\ &&+u[X_2, X_4]+v[X_3, X_4] -r[X_2, X_4]-s[X_3, X_4], \\ G_v&=&\left(u-v-\frac{2}{3}r+\frac{2}{3}s\right)X_2+\left(u+3v-\frac{2}{3}r-2s\right)X_3+[X_1, X_5]\\ &&+u[X_2, X_5]+v[X_3, X_5] -r[X_2, X_5]-s[X_3, X_5]. \end{eqnarray*} $

由相容性条件

$ \begin{eqnarray*} &G_{uv}=G_{vu}, \quad G_{up}=G_{pu}, \quad G_{uq}=G_{qu}, \quad G_{ur}=G_{ru}, \quad\quad G_{us}=G_{su}, \\ & G_{vp}=G_{pv}, \quad G_{vq}=G_{qv}, \quad G_{vr}=G_{rv}, \quad G_{vs}=G_{sv}, \quad G_{pq}=G_{qp}, \\ &G_{pr}=G_{rp}, \quad G_{ps}=G_{sp}, \quad G_{qr}=G_{rq}, \quad G_{qs}=G_{sq}, \quad G_{rs}=G_{sr}, \end{eqnarray*} $

我们有

$ [X_2, X_4]=-X_2+\frac{1}{3}X_3, \quad\quad \; [X_3, X_5]=\frac{1}{3}X_2-X_3, $
$ [X_3, X_4]=[X_2, X_5]=-\frac{1}{3}X_2-\frac{1}{3}X_3. $

经过一系列计算, 我们最终得到

$ \begin{eqnarray*} G&=&\left(u^2-\frac{1}{3}v^2+\frac{2}{3}uv-ur-\frac{1}{3}us -\frac{1}{3}vr+\frac{1}{3}vs\right)X_2\\ &&-\left(\frac{1}{3}u^2-v^2-\frac{2}{3}uv-\frac{1}{3}ur+\frac{1}{3}us +\frac{1}{3}vr+vs\right)X_3+pX_4+qX_5+X_6+uX_7+vX_8, \end{eqnarray*} $

其中$[X_1, X_4]=X_7, \quad [X_1, X_5]=X_8$.

$F, \, G$代入(3.8)式中的第六个方程, 我们就得到了所有$X_i, \; i=1, \cdots, 8$需满足的条件

$ [X_2, X_3]=0, \quad\quad [X_1, X_6]=0, \quad\quad [X_2, X_6]=-X_4, $
$ [X_3, X_6]=-X_5, \quad\quad [X_1, X_7]=X_4, \quad\quad [X_1, X_8]=X_5, $
$ [X_2, X_4]=-X_2+\frac{1}{3}X_3, \quad\quad [X_2, X_5]=[X_3, X_4]=-\frac{1}{3}X_2-\frac{1}{3}X_3, $
$ [X_3, X_5]=\frac{1}{3}X_2-X_3, \quad\quad [X_2, X_8]=[X_3, X_7]=-\frac{1}{3}[X_1, X_2]-\frac{1}{3}[X_1, X_3], $
$ [X_1, X_2]-\frac{1}{3}[X_1, X_3]+[X_2, X_7]=0, \quad\quad \frac{1}{3}[X_1, X_2]-[X_1, X_3]-[X_3, X_8]=0, $

这里所有的$X_i (i=1, 2, \cdots, 8)$均为待定矩阵.

如果我们设

$ \begin{eqnarray*} F&=&X_1+uX_2+vX_3-rX_2-sX_3+urX_4+vsX_5\\ &&-\frac{1}{2}u^2X_4 -\frac{1}{2}v^2X_5-\frac{1}{2}r^2X_4-\frac{1}{2}s^2X_5, \end{eqnarray*} $

同时

$ G_p=X_6, \quad\quad G_q=X_7, $

那么经过计算我们可得$X_4=X_5=0$, 此时便回到了前面所讨论的情形.

对于方程(3.1)我们引入下面的尺度对称

$ \begin{eqnarray}\label{e3.1.4} x \rightarrow x, \quad\quad t \rightarrow \lambda^{-1}t, \quad\quad u \rightarrow \lambda u, \quad\quad v \rightarrow \lambda v, \end{eqnarray} $ (3.12)

$X_i$满足

$ \begin{eqnarray}\label{e3.1.5}\begin{array}{ll} &X_1 \rightarrow X_1, \quad\quad X_2 \rightarrow \lambda^{-1}X_2, \quad\quad X_3 \rightarrow \lambda^{-1}X_3, \quad\quad X_4 \rightarrow X_4, \\ &X_5 \rightarrow X_5, \quad\quad X_6 \rightarrow \lambda X_6, \quad\quad X_7 \rightarrow X_7, \quad\quad X_8 \rightarrow X_8. \end{array}\end{eqnarray} $ (3.13)

同样利用Dodd和Fordy[23]的方法, 并结合Maple软件进行计算, 我们可得到

$ X_1=\left(\begin{array}{ccccc} 0&-\frac{1}{3\lambda}&0&0\\ -\frac{3}{4}\lambda &0&0&0\\ 0&\frac{1}{6\lambda}&0&-\frac{1}{3\lambda}\\ -\frac{3}{8}\lambda&0&-\frac{3}{4}\lambda &0 \end{array}\right), \quad X_2=\left(\begin{array}{ccccc} 0&~0~&0&~0\\ 1&0&0&~0\\ 0&0&0&~0\\ 0&0&1&~0 \end{array}\right), \quad X_3=\left(\begin{array}{ccccc} 0&~0~&0&~0\\ 1&0&0&~0\\ 0&0&0&~0\\ 1&0&1&~0 \end{array}\right), $
$ X_4=\left(\begin{array}{ccccc} -\frac{1}{3}&0&0&0\\ 0&\frac{1}{3}&0&0\\ \frac{1}{6}&0&-\frac{1}{3}&0\\ 0&-\frac{1}{6}&0&\frac{1}{3} \end{array}\right), \;X_5=\left(\begin{array}{ccccc} -\frac{1}{3}&0&0&0\\ 0&\frac{1}{3}&0&0\\ -\frac{1}{6}&0&-\frac{1}{3}&0\\ 0&\frac{1}{6}&0&\frac{1}{3} \end{array}\right), \; X_6=\left(\begin{array}{ccccc} 0&-\frac{1}{3}&0&0\\ -\frac{3}{4}\lambda^2&0&0&0\\ 0&\frac{1}{6}&0&-\frac{1}{3}\\ -\frac{3}{8}\lambda^2&0&-\frac{3}{4}\lambda^2&0 \end{array}\right), $
$ X_7=\left(\begin{array}{ccccc} 0&-\frac{2}{9\lambda}&0&0\\ \frac{1}{2}\lambda&0&0&0\\ 0&\frac{2}{9\lambda}&0&-\frac{2}{9\lambda}\\ 0&0&\frac{1}{2}\lambda&0 \end{array}\right), \quad X_8=\left(\begin{array}{ccccc} 0&-\frac{2}{9\lambda}&0&0\\ \frac{1}{2}\lambda&0&0&0\\ 0&0&0&-\frac{2}{9\lambda}\\ \frac{1}{2}\lambda&0& \frac{1}{2}\lambda&0 \end{array}\right). $

因此, 方程(3.1)的谱问题为

$ \begin{equation}\label{e3.9}\begin{array}{ll} \left(\begin{array}{ccccc}y^1 \\ y^2 \\ y^3 \\ y^4 \end{array}\right)_x =&\left(\begin{array}{ccccc} 0&-\frac{1}{3\lambda}&0&0\\ u+v-r-s-\frac{3}{4}\lambda &0&0&0\\ 0&\frac{1}{6\lambda}&0&-\frac{1}{3\lambda} \\ v-s-\frac{3}{8}\lambda&0& u+v-r-s-\frac{3}{4}\lambda&0 \end{array}\right) \left(\begin{array}{ccccc}y^1 \\ y^2 \\ y^3 \\ y^4 \end{array}\right), \\\\ \left(\begin{array}{ccccc}y^1 \\ y^2 \\ y^3 \\ y^4 \end{array}\right)_t =&\left(\begin{array}{ccccc} -\frac{1}{3}p-\frac{1}{3}q& -\frac{1}{9\lambda}(2u+2v+3\lambda)&0&0\\ A&\frac{1}{3}p+\frac{1}{3}q&0&0\\ \frac{1}{6}p-\frac{1}{6}q&\frac{1}{6}+\frac{2u}{9\lambda}& -\frac{1}{3}p-\frac{1}{3}q&-\frac{1}{9\lambda}(2u+2v+3\lambda)\\ B&-\frac{1}{6}p+\frac{1}{6}q&A&\frac{1}{3}p+\frac{1}{3}q \end{array}\right) \left(\begin{array}{ccccc}y^1 \\ y^2 \\ y^3 \\ y^4 \end{array}\right), \\ \end{array}\end{equation} $ (3.14)

这里$\lambda$为谱参数,

$ A=\frac{2}{3}(u+v)^2-\frac{2}{3}(u+v)(r+s)+\frac{1}{2}(u+v)\lambda-\frac{3}{4}\lambda^2, $
$ B=-\frac{1}{3}u^2+v^2+\frac{2}{3}uv+\frac{1}{3}ur-\frac{1}{3}us-\frac{1}{3}vr-vs+\frac{1}{2}v\lambda-\frac{3}{8}\lambda^2. $

方程(3.14)的相容性条件为

$ F_t-G_x+[F, G]=0, $

即广义耦合CH型方程(3.1).

4 第二个广义耦合CH型方程的延拓结构

文献[16]给出了下列复耦合KdV方程的延拓结构和Lax对表示

$ \begin{equation}\label{e4.1}\begin{array}{ll} u_t&=u_{xxx}+6uu_x+6vv_x, \\ v_t&=v_{xxx}+6u_xv+6uv_x, \end{array}\end{equation} $ (4.1)

$ \begin{equation}\label{e4.2}\begin{array}{ll} u_t&=u_{xxx}-3uu_x+3vv_x, \\ v_t&=v_{xxx}-3u_xv-3uv_x. \end{array}\end{equation} $ (4.2)

根据文献[25]我们可以知道方程(4.2)是完全可积的, 拥有双哈密顿结构, 递归算子和无穷多广义对称.方程(4.2)的哈密顿结构为

$ H_1=\int \Big(-\frac{1}{2}u_x^2-\frac{1}{2}v_x^2+u^3+v^3+2uv^2\Big) {\rm d}x, \quad\quad H_2=\int \frac{1}{2}(u^2+v^2) {\rm d}x, $
$ {\cal D}_2= \left(\begin{array}{ccccc} \partial &~~ 0\\ 0&~~\partial \end{array}\right), \quad\quad {\cal D}_1= \left(\begin{array}{ccccc} \partial^3+2\partial u+2u\partial ~~& 2\partial v+2v\partial\\ 2\partial v+2v\partial ~~& \partial^3+2\partial u+2u\partial \end{array}\right). $

利用三哈密顿求对偶系统的方法[1], 我们可得到(4.2)式的对偶系统为

$ \begin{equation}\label{e4.3}\begin{array}{ll} m_t=2mu_x+m_xu-2nv_x-n_xv, \\ n_t=2mv_x+m_xv+2nu_x+n_xu, \\ m=u-u_{xx}, \quad n=v-v_{xx}. \end{array}\end{equation} $ (4.3)

在本节中, 我们主要研究方程(4.3)的延拓结构.

利用独立变量(3.2)式, 方程(4.3)可被化为下列外微分2 -形式

$ \begin{equation}\label{e4.4}\begin{array}{ll} \alpha_1=&{\rm d}u\wedge {\rm d}t-p{\rm d}x\wedge {\rm d}t, \quad \alpha_2={\rm d}v\wedge {\rm d}t-q{\rm d}x\wedge {\rm d}t, \\ \alpha_3=&{\rm d}p\wedge {\rm d}t-r{\rm d}x\wedge {\rm d}t, \quad \alpha_4={\rm d}q\wedge {\rm d}t-s{\rm d}x\wedge {\rm d}t, \\ \alpha_5=&{\rm d}u\wedge {\rm d}x-{\rm d}r\wedge {\rm d}x+(3up-2pr-3vq+2qs){\rm d}x\wedge {\rm d}t-u{\rm d}r\wedge {\rm d}t+v{\rm d}s\wedge {\rm d}t, \\ \alpha_6=&{\rm d}v\wedge {\rm d}x-{\rm d}s\wedge {\rm d}x+(3vp+3uq-2qr-2ps){\rm d}x\wedge {\rm d}t-v{\rm d}r\wedge {\rm d}t-u{\rm d}s\wedge {\rm d}t. \end{array}\end{equation} $ (4.4)

易证理想$I=\{\alpha_1, \alpha_2, \alpha_3, \alpha_4, \alpha_5, \alpha_6\}$为闭理想.

$ \begin{equation}\label{e4.5}\begin{array}{ll} {\rm d}\alpha_1=&{\rm d}x\wedge \alpha_3, \qquad {\rm d}\alpha_2={\rm d}x\wedge \alpha_4, \\ {\rm d}\alpha_3=&{\rm d}x\wedge \alpha_1+{\rm d}t\wedge \alpha_5, \quad {\rm d}\alpha_4={\rm d}x\wedge \alpha_2+{\rm d}t\wedge \alpha_6, \\ {\rm d}\alpha_5=&-2p{\rm d}x\wedge \alpha_1+2q{\rm d}x\wedge \alpha_2-(3u-2r){\rm d}x\wedge \alpha_3+(3v-2s){\rm d}x\wedge \alpha_4\\ &+p{\rm d}t\wedge \alpha_5 -q{\rm d}t\wedge \alpha_6+{\rm d}r\wedge \alpha_1-{\rm d}s\wedge \alpha_2, \\ {\rm d}\alpha_6=&-2q{\rm d}x\wedge \alpha_1-2p{\rm d}x\wedge \alpha_2-(3v-2s){\rm d}x\wedge \alpha_3-(3u-2r){\rm d}x\wedge \alpha_4\\ &+q{\rm d}t\wedge \alpha_5 +p{\rm d}t\wedge \alpha_6+{\rm d}r\wedge \alpha_2+{\rm d}s\wedge \alpha_1. \end{array}\end{equation} $ (4.5)

同第三节中的步骤, 方程(4.3)等价于下述一系列非线性偏微分方程

$ \begin{equation}\label{e4.6}\begin{array}{ll} &F_p=F_q=0, \quad F_u+F_r=0, \qquad F_v+F_s=0, \\ & G_r=-uF_u-vF_v, G_s=vF_u-uF_v, \\ &pG_u+qG_v+rG_p+sG_q-(3up-2pr-3vq+2qs)F_u\\ &-(3vp+3uq-2qr-2ps)F_v-[F, G]=0. \end{array}\end{equation} $ (4.6)

经过计算与分析, 我们可得

$ F=X_1+uX_2+vX_3-rX_2-sX_3, $
$ G=(u^2-v^2-ur+vs)X_2+(2uv-us-vr)X_3+pX_4+qX_5+X_6+uX_7+vX_8, $

并且$X_i, \; i=1, \cdots, 8$满足

$ [X_2, X_3]=0, \quad [X_1, X_6]=0, \quad [X_2, X_6]=-X_4, \quad [X_3, X_6]=-X_5, $
$ [X_1, X_4]=X_7, \quad [X_1, X_7]=X_4, \quad [X_1, X_5]=X_8, \quad [X_1, X_8]=X_5, $
$ [X_2, X_4]=-X_2, \quad [X_3, X_5]=X_2, \quad [X_2, X_5]=[X_3, X_4]=-X_3, $
$ [X_2, X_8]=[X_3, X_7]=-[X_1, X_3], \quad [X_1, X_2]+[X_2, X_7]=0, \quad [X_1, X_2]-[X_3, X_8]=0, $

其中$X_i \, (i=1, 2, \cdots, 8)$均为待定矩阵.

我们可以看到

$ [X_4, X_2]=X_2, \quad\quad [X_6, X_2]=X_4, $

由定理2.1可得

$ X_2=e_-, \quad\quad X_4=-\frac{1}{2}h. $

首先考虑将延拓代数$\{X_i, \, i=1, 2, \cdots, 8 \}$嵌入到Lie代数${\rm sl}(2, {\Bbb C})$中.

结合同样的尺度对称(3.12)及(3.13), 我们可以求得

$ X_1=\left(\begin{array}{ccccc} 0&\frac{\sqrt{2}i}{4\lambda}\\ -\frac{\sqrt{2}\lambda i}{2} &0 \end{array}\right), \qquad X_2=\left(\begin{array}{ccccc} 0&~~0\\ 1&~~0 \end{array}\right), \qquad X_3=\left(\begin{array}{ccccc} 0&~~0\\ i&~~0 \end{array}\right), $
$ X_4=\left(\begin{array}{ccccc} -\frac{1}{2}&~~0\\ 0&~~\frac{1}{2} \end{array}\right), \;\;\;X_5=\left(\begin{array}{ccccc} -\frac{1}{2}i&0\\ 0&\frac{1}{2}i \end{array}\right), \qquad X_6=\left(\begin{array}{ccccc} 0&-\frac{1}{2}\\ \lambda^{2}&0 \end{array}\right), $
$ X_7=\left(\begin{array}{ccccc} 0& \frac{\sqrt{2}i}{4\lambda}\\ \frac{\sqrt{2}\lambda i}{2} &0 \end{array}\right), \qquad X_8=\left(\begin{array}{ccccc} 0& -\frac{\sqrt{2}}{4\lambda}\\ -\frac{\sqrt{2}\lambda}{2} &0 \end{array}\right). $

因此

$ F=\left(\begin{array}{ccccc} 0& ~~\frac{\sqrt{2}i}{4\lambda}\\ u-r+(v-s)i-\frac{\sqrt{2}\lambda i}{2}&~~0 \end{array}\right), $
$ G=\left(\begin{array}{ccccc} -\frac{p}{2}-\frac{q}{2}i&~~ -\frac{1}{2}+\frac{\sqrt{2}i}{4\lambda}(u+vi) \\ A+Bi& ~~ \frac{p}{2}+\frac{q}{2}i \end{array}\right), $

这里$\lambda$为谱参数,

$ A=u^2-v^2-ur+vs+\lambda^2, \quad B=2uv-us-vr+\frac{\sqrt{2}}{2}\lambda(u+vi), \quad i^2=-1. $

Lax对的矩阵表示形式为

$ y_x=Fy, \quad\quad y_t=Gy, \quad \; y=(y^1, y^2)^{\rm T}. $

因此, 我们可得到方程(4.3)的线性谱问题为

$ \begin{equation}\label{e4.7}\begin{array}{ll} \left(\begin{array}{ccccc}y^1 \\ y^2 \end{array}\right)_x =&\left(\begin{array}{ccccc} 0& ~~\frac{\sqrt{2}i}{4\lambda}\\ u-r+(v-s)i-\frac{\sqrt{2}\lambda i}{2} &~~ 0 \end{array}\right) \left(\begin{array}{ccccc}y^1 \\ y^2 \end{array}\right), \\[8mm] \left(\begin{array}{ccccc}y^1 \\ y^2 \end{array}\right)_t =&\left(\begin{array}{ccccc} -\frac{p}{2}-\frac{q}{2}i &~~ -\frac{1}{2}+\frac{\sqrt{2}i}{4\lambda}(u+vi) \\ A+Bi &~~ \frac{p}{2}+\frac{q}{2}i \end{array}\right) \left(\begin{array}{ccccc}y^1 \\ y^2 \end{array}\right). \end{array}\end{equation} $ (4.7)

$w=u+vi$, 则上述谱问题等价于

$ \begin{equation}\label{e4.8}\begin{array}{ll} \left(\begin{array}{ccccc}y^1 \\ y^2 \end{array}\right)_x =&\left(\begin{array}{ccccc} 0&~~~~\frac{\sqrt{2}i}{4\lambda}\\ w-w_{xx}-\frac{\sqrt{2}\lambda i}{2} &~~~~ 0 \end{array}\right) \left(\begin{array}{ccccc}y^1 \\ y^2 \end{array}\right), \\[8mm] \left(\begin{array}{ccccc}y^1 \\ y^2 \end{array}\right)_t =&\left(\begin{array}{ccccc} -\frac{1}{2}w_x & ~~ -\frac{1}{2}+\frac{\sqrt{2}i}{4\lambda}w \\ w(w-w_{xx})+\frac{\sqrt{2}}{2}w\lambda i+\lambda^2& ~~ \frac{1}{2}w_x \end{array}\right) \left(\begin{array}{ccccc}y^1 \\ y^2 \end{array}\right). \end{array}\end{equation} $ (4.8)

方程(4.7)或(4.8)的相容性条件为

$ F_t-G_x+[F, G]=0, $

即广义耦合复CH型方程(4.3).

参考文献
[1] Olver P J, Rosenau P. Tri-Hamiltonian duality between solitons and solitary-wave solutions having compact support. Physical Review E, 1996, 53(2): 1900–1906. DOI:10.1103/PhysRevE.53.1900
[2] Wahlquist H D, Estabrook F B. Prolongation structures of nonlinear evolution equations. J Math Phys, 1975, 16(1): 1–7.
[3] Estabrook F B, Wahlquist H D. Prolongation structures of nonlinear evolution equations Ⅱ. J Math Phys, 1976, 17(7): 1293–1297. DOI:10.1063/1.523056
[4] Roelofs G H M, Martini R. Prolongation structure of the KdV equation in the bilinear form of Hirota. Journal of Physics A:Mathematical and General, 1990, 23(11): 1877–1884. DOI:10.1088/0305-4470/23/11/015
[5] Morris H C. Prolongation structures and nonlinear evolution equations in two spatial dimensions Ⅱ:A generalized nonlinear Schrödinger equation. J Math Phys, 1977, 18(2): 285–288. DOI:10.1063/1.523248
[6] Duan X J, Deng M, Zhao W Z, et al. The prolongation structure of the inhomogeneous equation of the reaction-diffusion type. Journal of Physics A:Mathematical and Theoretical, 2007, 40(14): 3831–3837. DOI:10.1088/1751-8113/40/14/006
[7] Yang Y Q, Chen Y. Prolongation structure of the equation studied by Qiao. Communications in Theoretical Physics, 2011, 56(3): 463–466. DOI:10.1088/0253-6102/56/3/13
[8] Wang D S, Ma Y Q, Li X G. Prolongation structures and matter-wave solitons in F=1 spinor BoseEinstein condensate with time-dependent atomic scattering lengths in an expulsive harmonic potential. Communications in Nonlinear Science and Numerical Simulation, 2014, 19(10): 3556–3569. DOI:10.1016/j.cnsns.2014.02.019
[9] Bracken P. Exterior differential systems prolongations and application to a study of two nonlinear partial differential equations. Acta Appl Math, 2011, 113(3): 247–263. DOI:10.1007/s10440-010-9597-z
[10] Stalin S, Senthilvelan M. A note on the prolongation structure of the cubically nonlinear integrable Camassa-Holm type equation. Physics Letters A, 2011, 375(43): 3786–3788. DOI:10.1016/j.physleta.2011.08.057
[11] 白永强, 裴明, 高文娟. Camassa-Holm和Degasperis-Procesi方程的延拓结构(英文). 河南大学学报(自然科学版), 2011, 41(6): 551–555.
Bai Y Q, Pei M, Gao W J. Prolongation structures of the Camassa-Holm and Degasperis-Procesi equations. Journal of Henan University (Natural Science), 2011, 41(6): 551–555. DOI:10.3969/j.issn.1003-4978.2011.06.001
[12] Bracken P. An exterior differential system for a generalized Korteweg-de Vries equation and its associated integrability. Acta Appl Math, 2007, 95(3): 223–231. DOI:10.1007/s10440-007-9086-1
[13] Wang D S. Integrability of a coupled KdV system:Painlevé property, Lax pair and Bäcklund transformation. Applied Mathematics and Computation, 2010, 216(4): 1349–1354. DOI:10.1016/j.amc.2010.02.030
[14] Wang D S. Complete integrability and the Miura transformation of a coupled KdV equation. Applied Mathematics Letters, 2010, 23(6): 665–669. DOI:10.1016/j.aml.2010.02.002
[15] Cao Y H, Wang D S. Prolongation structures of a generalized coupled Korteweg-de Vries equation and Miura transformation. Communications in Nonlinear Science and Numerical Simulation, 2010, 15(9): 2344–2349. DOI:10.1016/j.cnsns.2009.10.020
[16] Wang D S. Integrability of the coupled KdV equations derived from two-layer fluids:Prolongation structures and Miura transformations. Nonlinear Analysis:Theory, Methods and Applications, 2010, 73(1): 270–281. DOI:10.1016/j.na.2010.03.021
[17] Wang D S, Wei X Q. Integrability and exact solutions of a two-component Korteweg-de Vries system. Applied Mathematics Letters, 2016, 51: 60–67. DOI:10.1016/j.aml.2015.07.007
[18] Wang D S, Liu J, Zhang Z F. Integrability and equivalence relationships of six integrable coupled Kortewegde Vries equations. Mathematical Methods in the Applied Sciences, 2016, 39(12): 3516–3530. DOI:10.1002/mma.v39.12
[19] Humphreys J E. Introduction to Lie Algebras and Representation Theory. New York: Springer-Verlag, 1972.
[20] Jacobson N. Lie Algebras. New York: Dover Publications, 1979.
[21] Foursov M V. On integrable coupled KdV-type systems. Inverse Problems, 2000, 16(1): 259–274. DOI:10.1088/0266-5611/16/1/319
[22] Flanders H. Differential Forms. New York: Academic Press, 1963: 1-73.
[23] Dodd R, Fordy A P. The prolongation structures of quasi-polynomial flows. Proceedings of the Royal Society of London A:Mathematical, Physical and Engineering Sciences, The Royal Society, 1983, 385(1789): 389–429. DOI:10.1098/rspa.1983.0020
[24] 罗婷. 多分量KdV, mKdV方程组的对偶系统[D]. 西安: 西北大学, 2013
Luo T. Dual System of Multi-Component KdV and mKdV Equations[D]. Xi'an: Northwest Univesity, 2013
[25] Fuchssteiner B. The Lie algebra structure of degenerate Hamiltonian and bi-Hamiltonian systems. Progress of Theoretical Physics, 1982, 68(4): 1082–1104. DOI:10.1143/PTP.68.1082