数学物理方法 12 非线性方程

最新推荐文章于 2022-01-04 16:52:30 发布

longji

最新推荐文章于 2022-01-04 16:52:30 发布

阅读量792

点赞数 2

分类专栏：数学物理方法文章标签：数学物理方法

本文链接：https://blog.csdn.net/longji/article/details/79426186

版权

数学物理方法专栏收录该内容

16 篇文章 18 订阅

订阅专栏

$\color{blue}{第十二章非线性方程}$

$\color{blue}{12.0 引言}$

$\color{blue}{12.0.1 为何要引入非线性方程}$

$1.线性,只是理想的状态,初步的近似,大多事物的本来面目,却是非线性的.$
$2.现代科学技术的发展,使非线性科学占有及其重要的地位.$

$\color{blue}{12.0.2 何谓非线性方程}$

$方程中含有未知函数和未知函数偏导数的高次项的方程.$
$如,KdV: u_{\tau} + u_{\xi} + 12 uu_{\xi} + u_{\xi \xi \xi} = 0 \quad (1)$

$\color{blue}{12.0.3 非线性方程的特点}$

$1.不满足叠加原理$
$2.其定解问题一般不满足适定性;$
$3.没有普遍的理论和求解方法,大多都不能求得其解析解;$
$4.其解对初始条件具有敏感性(如,蝴蝶效应);$
$5.其求解途径为:$
$(1)解依赖于自变量的\left \lbrace \begin{array}{l}幂组合\left \lbrace \begin{array}{l} u = u(\xi) \\ \xi = a^{\alpha} t^{\beta} \end{array} \right. \\ 线性组合\left \lbrace \begin{array}{l}u= u(\xi) \\ \xi = x + at \end{array} \right. \end{array} \right. 将偏微分方程化为常微分方程求解$
$(2)通过自变量或函数变换,将非线性方程化为线性方程解$

$大气对流的洛伦兹(Lorenz)方程为$
$\left \lbrace \begin{array}{l} \dot x(t) = -10x + 10 y \\ \dot y(t) = 28 x - y - xz \\ \dot z(t) = \dfrac{8}{3} x + xy \end{array} \right.$

$\color{blue}{12.1 非线性方程的某些初等解法}$

$\color{blue}{12.1.1 Kirchhoff 变换}$

$对于\nabla \cdot [G(u) \nabla u] = 0 \quad (1)$
$若选 w = \int_{u_0}^{u} G(\xi) d \xi \quad (2) - Kirchhoff变换$
$则(1) \to \Delta w = 0$
$证明:令w = w(u), 使\nabla w = G(u) \nabla u, 则(1) \to \Delta w = 0$
$此时, \dfrac{dw}{du} \nabla u = G(u) \nabla u \to \dfrac{dw}{du} = G(u)$
$即: w = \int_{u_0}^{u} G(\xi) d \xi \quad (2)$

$例题:求解:\left \lbrace \begin{array}{l}\nabla \cdot [u^2 \nabla u] = 0 \\ u|_{y=0} = Ax \end{array} \right. \quad (3)$
$选:w = \int_0^u \xi^2 d \xi = \dfrac{1}{3} u^3, 则(3) \to \left \lbrace \begin{array}{l}\Delta w = 0 \\ w |_{y=0} = \dfrac{1}{3}A^3 x^3 \end{array} \right.$

$\color{blue}{12.1.2 Cole-Hopf变换}$

$对于u_t + uu_x = \delta u_{xx} \quad (4) - Burgers方程$
$若作变换 u = -2\delta \dfrac{\partial \ln v}{\partial x} \quad (5) - Cole-Hopf变换$
$则(4) \to v_t = \lambda v_{xx}$
$证明:1.求解(4)即要求解$
$\psi_t + \dfrac{1}{2}\psi_x^2 = \delta \psi_{xx} \quad (6)$
$其中, \psi_x = u \quad (7)$
$\because (4) \to \dfrac{\partial u}{\partial t} = \dfrac{\partial}{\partial x}(\delta u_x - \dfrac{u^2}{2})$
$由全微分存在的充要条件有:$
$d \psi = u dx + (\delta u_x - \dfrac{u^2}{2}) dt$
$其中\left \lbrace \begin{array}{l}\psi_x = u \\ \psi_t = \delta u_x - \dfrac{u^2}{2} = \delta \psi_{xx} - \dfrac{1}{2} \psi_x^2 \end{array} \right.$
$即(4) \to \psi_t + \dfrac{1}{2} \psi_x^2 = \delta \psi_{xx} \quad (8)$
$2.引入变换v = g(\psi),使(8) \to v_t = \delta v_{xx} \quad (9)$
$令 v = g(\psi), 则 v_t = g^{\prime}(\psi) \psi_t, v_x = g^{\prime}(\psi)\psi_x$
$v_{xx} = g^{\prime \prime}(\psi)(\psi_{x})^2 + g^{\prime}(\psi)\psi_{xx}$
$于是(9) \to \psi_t - \delta \dfrac{\psi^{\prime \prime}(\psi)}{g^{\prime}(\psi)}(\psi_x)^2 = \delta \psi_{xx} \quad (10)$
$对比(8)和(10): -\delta \dfrac{g^{\prime \prime}(\psi)}{g^{\prime}(\psi)} = \dfrac{1}{2} \to g^{\prime \prime}(\psi) = - \dfrac{1}{2 \delta} g^{\prime}(\psi)$
$\to g(\psi) = c_1 e^{-\frac{1}{2 \delta}\psi} + c_2$
$取c_1 = 1, c_2 = 0 \to g(\psi) = e^{-\frac{1}{2\delta}\psi} = v$
$\to \psi = -2\delta \ln v,$

$3.结论:故作变换u = \psi_x = -2\delta \dfrac{\partial \ln v}{\partial x}$
$u_t + uu_{x} = \delta u_{xx} \quad (4) \to v_t = \lambda v_{xx}$

$\color{blue}{12.1.3 相似变换}$

$对于\dfrac{\partial u}{\partial t} = \dfrac{\partial}{\partial x}[G(u) \dfrac{\partial u}{\partial x}] \quad (11)$
$作变换\left \lbrace \begin{array}{l} u = u(\xi) \quad (12) \\ \xi = \dfrac{x}{\sqrt{t}} \quad (13) -Boltzman变换 \end{array} \right.$
$而称:\left \lbrace \begin{array}{l}u = u(\xi) \quad (12) \\ \xi = x^{\alpha} t^{\beta} \quad (14) -相似变换 \end{array} \right. -自型解$
$证明:对于\dfrac{\partial u}{\partial t} = \dfrac{\partial}{\partial x}[G(u)\dfrac{\partial u}{\partial x}] \quad (11)$
$令\left \lbrace \begin{array}{l}u = u(\xi) \quad (12) \\ \xi = x^{\alpha} t^{\beta} \quad (14) \end{array} \right. 则$
$\dfrac{\partial u}{\partial t} = \dfrac{du}{d\xi} \dfrac{\partial \xi}{\partial t} = \beta x^{\alpha} t^{\beta - 1} u^{\prime}(\xi), \dfrac{\partial u}{\partial x} = \alpha x^{\alpha - 1} t^{\beta} u^{\prime}(\xi)$
$\dfrac{\partial}{\partial x}[G(u)\dfrac{\partial u}{\partial x}] = G(u) \dfrac{\partial}{\partial x}\dfrac{\partial u}{\partial x} +(\dfrac{\partial u}{\partial x})^2 G^{\prime}(u)$
$= \alpha(\alpha - 1)x^{\alpha - 2} t^{\beta} G(u) u^{\prime}(\xi) + \alpha^2 x^{2(\alpha - 1)} t^{2 \beta} \dfrac{d}{d \xi}[G(u)u^{\prime}(\xi)]$
$(11) \to \beta\dfrac{x^2}{t}\xi\dfrac{du}{d\xi} = \alpha(\alpha-1)\xi G(u) \dfrac{du}{d\xi} + \alpha^2 \xi^2 \dfrac{d}{d\xi}[G(u)\dfrac{du}{d\xi}] \quad (11)^{\prime}$
$选\dfrac{x^2}{t} = g(\xi) = \xi^2, 即\xi = \dfrac{x}{\sqrt{t}},$
$亦即(14)中选\alpha = 1, \beta = -\dfrac{1}{2}$
$(11)^{\prime} \to -\dfrac{\xi}{2}\dfrac{du}{d\xi} = \dfrac{d}{d\xi}[G(u)\dfrac{du}{d\xi}]$
$(11) \to \dfrac{d}{d\xi}[G(u)\dfrac{du}{d\xi}] + \dfrac{\xi}{2}\dfrac{du}{d\xi} = 0$

$\color{blue}{12.1.4 行波解}$

$对于\dfrac{\partial u}{\partial t} = \dfrac{\partial}{\partial x}[u^n \dfrac{\partial u}{\partial x}] \quad (15)$
$令\left \lbrace \begin{array}{l} u = g(\xi) \quad (16) \\ \xi = x + at \quad (17) \end{array} \right. -行波解$
$则\dfrac{\partial u}{\partial t} = a \dfrac{dg}{d\xi}, \dfrac{\partial u}{\partial x} = \dfrac{dg}{d\xi}$
$(15) \to a\dfrac{dg}{d\xi} = \dfrac{d}{d\xi}[g^n\dfrac{dg}{d\xi}]$
$\to u = g(\xi) = \lbrace n[a(x + at) + c] \rbrace^{\frac{1}{n}}$

$\color{blue}{12.1.5 端迹变换}$

$\left \lbrace \begin{array}{l}u_1\dfrac{\partial F}{\partial x} + u_2\dfrac{\partial F}{\partial y} + u_3\dfrac{\partial E}{\partial x} + u_4\dfrac{\partial E}{\partial y} = 0 \\ v_1\dfrac{\partial F}{\partial x} + v_2\dfrac{\partial F}{\partial y} + v_3\dfrac{\partial E}{\partial x} + v_4\dfrac{\partial E}{\partial y} = 0 \end{array} \right.(16)$
$证明:令J = \dfrac{\partial (F, E)}{\partial (x, y)} = F_xE_y - E_xF_y \neq 0 \quad (17)$
$则可引入\left \lbrace \begin{array}{l}x = x(F, E) \\ y = y(F, E) \end{array} \right. (18)$
$\dfrac{\partial}{\partial x}(18): \left \lbrace \begin{array}{l}1 = x_FF_x + x_EE_x \\ 0 = y_FF_x + y_EE_x \end{array} \right.$
$\to \left \lbrace \begin{array}{l}F_x = \dfrac{y_E}{x_Ey_E - x_Ey_F} = \dfrac{y_E}{j_E} = Jy_E \\E_x = -Jy_F \end{array} \right. (19)$
$j = \dfrac{\partial (x, y)}{\partial (F, E)} = J^{-1} \neq 0$
$\dfrac{\partial}{\partial y}(18):\left \lbrace \begin{array}{l}F_y = -Jx_E \\ E_y = Jx_F \end{array} \right.(20)$
$(19),(20)代入(16):$
$\left \lbrace \begin{array}{l}u_1\dfrac{\partial y}{\partial E} - u_2\dfrac{\partial x}{\partial E} - u_3\dfrac{\partial y}{\partial F} + u_4\dfrac{\partial x}{\partial F} = 0 \\ v_1 \dfrac{\partial y}{\partial E} - v_2\dfrac{\partial x}{\partial E} - v_3\dfrac{\partial y}{\partial F} + v_4\dfrac{\partial x}{\partial F} = 0 \end{array} \right.$

$\color{blue}{12.1.6 小结}$

$1.\nabla \cdot[G(u)\nabla u] = 0 \to \Delta w = 0,若选w = \int_{u_0}^{u}G(\xi)d\xi -Kirchhoff变换$
$2.对于u_t + uu_x = \delta u_{xx} \to v_t = \lambda v_{xx}$
$若作变换u = -2\delta \dfrac{\partial \ln v}{\partial x} - Cole-Hopf变换$
$3.\dfrac{\partial u}{\partial t} = \dfrac{\partial}{\partial x}[G(u)\dfrac{\partial u}{\partial x}] \to \dfrac{d}{d\xi}[G(u)\dfrac{du}{d\xi}] + \dfrac{\xi}{2}\dfrac{du}{d\xi} = 0$
$若令u = u(\xi), \xi = \dfrac{x}{\sqrt{t}} -相似变换$
$4.\dfrac{\partial u}{\partial t} = \dfrac{\partial}{\partial x}[u^n\dfrac{\partial u}{\partial x}] \to u = g(\xi) = \lbrace n[a(x+at) + c]\rbrace ^{\frac{1}{n}}$
$[u = g(\xi), \xi = x + at] -行波法$
$5.\left \lbrace \begin{array}{l}u_1\dfrac{\partial F}{\partial x} + u_2\dfrac{\partial F}{\partial y} + u_3\dfrac{\partial E}{\partial x} + u_4\dfrac{\partial E}{\partial y} = 0 \\ v_1\dfrac{\partial F}{\partial x} + v_2\dfrac{\partial F}{\partial y} + v_3\dfrac{\partial E}{\partial x} + v_4\dfrac{\partial E}{\partial y} = 0 \end{array} \right.$
$x = x(F, E), y = (F, E) -端迹变换$
$\to \left \lbrace \begin{array}{l}u_1\dfrac{\partial y}{\partial E} - u_2\dfrac{\partial x}{\partial E} - u_3\dfrac{\partial y}{\partial F} + u_4\dfrac{\partial x}{\partial F} = 0 \\ v_1 \dfrac{\partial y}{\partial E} - v_2\dfrac{\partial x}{\partial E} - v_3\dfrac{\partial y}{\partial F} + v_4\dfrac{\partial x}{\partial F} = 0 \end{array} \right.$

$\color{blue}{\S 12.2 孤波和孤子}$

$\color{blue}{12.2.1 KdV方程}$

$1.孤波的起源:$
$孤波:单峰行进,常速,波形不变.$
1895年:荷兰的Kortweg & de Vries观察浅水沟中水波,总结得KdV方程:
$u_{tau} + u_{\xi} + 12uu_{\xi} + u_{\xi \xi \xi} = 0$

$2.孤波解:$
$u_{tau} + u_{\xi} + 12uu_{\xi} + u_{\xi \xi \xi} = 0 \quad (1)$
$令\left \lbrace \begin{array}{l}u = u(\theta) \quad (2) \\ \theta = a\xi - \omega \tau + \delta \quad (3) \end{array} \right.$
$其中,a - 常数, \delta - 相位因子, u^{(n)} \stackrel{|\theta| \to \infty}{--\to} 0(n=1, 2, \cdots)$
$则u_{\tau} = -\omega u_{\theta}, u_{\xi} = au_{\theta}, u_{\xi \xi \xi} = a^3u_{\theta \theta \theta},$
$-\omega u_{\theta} + au_{\theta} + 12auu_{\theta} + a^3 u_{\theta \theta \theta} = 0$
$(1) \to u_{\theta \theta \theta} + \dfrac{12}{a^2}uu_{\theta} - \dfrac{\omega - a}{a^3} u_{\theta} = 0$
$取\omega = a + a^3, (1) \to u_{\theta \theta \theta} + \dfrac{12}{a^2}uu_{\theta} - u_{\theta} = 0 \quad (4)$
$\int (4)d \theta: \int \dfrac{d}{d\theta} u_{\theta \theta} d \theta + \dfrac{12}{a^2}\int u du - \int du = 0$
$即u_{\theta \theta} + \dfrac{6}{a^2} u^2 - u + c_1 = 0 \quad (5)$
$\int (5) \cdot u_{\theta} d\theta: \int u_{\theta} du_{\theta} + \dfrac{6}{a^2} \int u^2 du - \int u du + c_1 \int du = 0$
$\dfrac{1}{2}u_{\theta}^2 + \dfrac{2}{a^2}u^3 - \dfrac{1}{2}u^2 + c_1 u + c_2 = 0$
$使u^{(n)} \stackrel{|\theta| \to \infty}{--\to} 0(n = 0, 1, 2, \cdots) \to c_1 = c_2 = 0$
$\to u_{\theta}^2 = u^2(1 - \dfrac{4}{a^2}u) \to u_0 = \dfrac{u}{a}\sqrt{a^2 - 4u}$
$\to \dfrac{adu}{u\sqrt{a^2 - 4u}} = d\theta \to \theta = \ln \dfrac{a - \sqrt{a^2 - 4u}}{a + \sqrt{a^2 - 4u}}$
$\to e^{\theta} = \dfrac{a - \sqrt{a^2 - 4u}}{a + \sqrt{a^2 - 4u}} = \dfrac{a^2 - 2u - a\sqrt{a^2 - 4u}}{2u}$
$\to u = \dfrac{a^2 e^{\theta}}{(e^{\theta} + 1)^2} = \dfrac{a^2}{e^{\theta} + e^{-\theta} + 2}$
$=\dfrac{a^2}{(e^{\frac{\theta}{2}} + e^{-\frac{\theta}{2}})^2} = \dfrac{a^2}{4} sech^2 \dfrac{\theta}{2}$
$\color{blue}{\therefore u(\xi, \tau) = \dfrac{a^2}{4} sech^2 \dfrac{a}{2}[\xi - (1 + a^2)\tau + \dfrac{\delta}{a}]}$
$常速: 1 + a^2,振幅:a^2/4$

$\color{blue}{12.2.2 正弦Gordon方程}$

$\Phi_{xx} - \Phi_{tt} = \sin \Phi \quad (6)$
$1.引入变量代换:\xi = \dfrac{x-t}{2}, \tau = \dfrac{x + t}{2},$
$(6) \to \Phi_{\xi \tau} = \sin \Phi \quad (7)$
$令(3)的两线性无关的解为:$
$\left \lbrace \begin{array}{l}\Phi = u(\xi, \tau) + v(\xi, \tau) \quad (8) \\ \bar{\Phi} = u(\xi, \tau) - v(\xi, \tau) \quad (9) \end{array} \right.$
$令\left \lbrace \begin{array}{l}u_{\xi} = f(v) \quad (10) \\ v_{\tau} = g(u) \quad (11) \end{array} \right.$

$2.确定g(u), f(v):$
$\left \lbrace \begin{array}{l}(10) \\ (11) \end{array} \right. \to \left \lbrace \begin{array}{l}u_{\xi \tau} = g(u)f^{\prime}(v) \quad (12) \\ v_{\xi \tau} = f(v)g^{\prime}(u) \quad (13) \end{array} \right.$
$(12)+(13):(u+v)_{\xi \tau} = g(u)f^{\prime}(v) + g^{\prime}(u)f(v)$
$(12)-(13):(u-v)_{\xi \tau} = g(u)f^{\prime}(v) - g^{\prime}(u)f(v)$
$\sin(u+v) = g(u)f^{\prime}(v) + g^{\prime}(u)f(v) \quad (14)$
$\sin(u-v) = g(u)f^{\prime}(v) - g^{\prime}(u)f(v) \quad (15)$
$(14)+(15): g(u)f^{\prime}(v) = \sin u \cos v$
$\to \dfrac{g(u)}{\sin u} = \dfrac{\cos v}{f^{\prime}(v)} = \alpha \to g(u) = \alpha \sin u \quad (16)$
$(14)-(15):g^{\prime}(u)f(v) = \sin v \cos u$
$\to \dfrac{f(v)}{\sin v} = \dfrac{\cos u}{g^{\prime}(u)} = \beta \to f(v) = \beta \sin v = \dfrac{1}{\alpha} \sin v \quad (17)$

$3.引入Baklund变换$
$(8)+(9): u = \dfrac{1}{2}(\Phi + \bar \Phi) \quad (18)$
$(8)-(9): v = \dfrac{1}{2}(\Phi - \bar \Phi) \quad (19)$
$故由(18),(10),(17) \to \dfrac{1}{2}(\Phi + \bar \Phi)_{\xi} = \dfrac{1}{\alpha} \sin \dfrac{\Phi - \bar \Phi}{2} \quad (20)$
$而由(19),(11),(16) \to \dfrac{1}{2}(\Phi - \bar \Phi)_{\tau} = \alpha \sin \dfrac{\Phi + \bar \Phi}{2} \quad (21)$
$-巴古朗德(Baklund)变换$

$4.正弦Gordon方程的解$
$由\bar \Phi = 0有(20) \to \Phi_{\xi} = 2 \dfrac{1}{\alpha} \sin \dfrac{\Phi}{2} \quad (22)$
$(21) \to \Phi_{\tau} = 2 \alpha \sin \dfrac{\Phi}{2} \quad (23)$
$(22) \to \int \csc \dfrac{\Phi}{2}d\dfrac{\Phi}{2} = \dfrac{1}{\alpha}\int d \xi \to \dfrac{1}{\alpha} \xi = \ln \tan \dfrac{\Phi}{4} - c_1(\tau)$
$\to \Phi = 4 \tan^{-1} exp[\dfrac{1}{\alpha} \xi + c_1(\tau)]$
$(23) \to \int \csc \dfrac{\Phi}{2}d \dfrac{\Phi}{2} = 2\alpha \int d\tau \to \alpha \tau = \ln \tan \dfrac{\Phi}{4} - c_2(\xi)$
$\to \left \lbrace \begin{array}{l}c_1(\tau) = \alpha \tau + \delta \\ c_2(\xi) = \dfrac{1}{\alpha} \xi + \delta \end{array} \right.$
$\Phi = 4 \tan^{-1} exp[\dfrac{1}{\alpha} \xi + \alpha \tau + \delta]$
$\color{blue}{\Phi = 4 \tan^{-1} exp[a(x-bt) + \delta] - 扭结(孤波解)}$
$a = \dfrac{\alpha^2 + 1}{2 \alpha}, b = \dfrac{1 - \alpha^2}{1 + \alpha^2}$
$\Phi_{\xi} = 4 \dfrac{1}{\alpha} sech[a(x-bt) + \delta]$
$\Phi_{\tau} = 4 \alpha sech[a(x-bt) + \delta]$

$\color{blue}{12.2.3 非线性薛定谔方程}$

$i\Phi_t + \Phi_{xx} + \beta \Phi \bar \Phi^2 = 0 \quad (25)$
$令\left \lbrace \begin{array}{l}\Phi = e^{i(kx-vt)}u(\theta) \\ \theta = x-bt \end{array} \right. (26)$
$(25) \to u_{\theta \theta} + i(2k - b) u_{\theta} + (v - k^2) u + \beta u^3 = 0$
$选k = \dfrac{b}{2}, v = \dfrac{b^2}{4} - a^2 \to u_{\theta \theta} - a^2 u + \beta u^3 = 0 \quad (27)$
$\int (27) \cdot u_{\theta} d \theta : u_{\theta}^2 = a^2 u^2 - \dfrac{\beta}{2} u^4 + c$
$由|x| \to \infty, \Phi^{(n)}(x) \to 0,有c = 0$
$\dfrac{du}{u\sqrt{a^2 - \dfrac{\beta}{2}u^2}} = d \theta \to \theta = -\dfrac{1}{a} \ln \dfrac{a + \sqrt{a^2 - \dfrac{\beta}{2}u^2}}{u\sqrt{\dfrac{\beta}{2}}}$
$i\Phi_t + \Phi_{xx} + \beta \Phi \bar \Phi^2 = 0 \quad (25)$
$e^{-a \theta} = \dfrac{a + \sqrt{a^2 - \dfrac{\beta}{2}u^2}}{u\sqrt{\dfrac{\beta}{2}}} \to u(\theta) = a \sqrt{\dfrac{2}{\beta}} sech a \theta$

$\color{blue}{\varphi(x, t) = a\sqrt{\dfrac{2}{\beta}} exp\lbrace i[\dfrac{1}{2}bx - (\dfrac{1}{4}b^2 - a^2)t]\rbrace sech[a(x-bt)] - 孤波解}$

$\color{blue}{12.2.4 小结}$

$1.KdV方程:$
$u_{\tau} + u_{\xi} + 12uu_{\xi} + u_{\xi \xi \xi} = 0 \quad (1)$
$令\left \lbrace \begin{array}{l}u = u(\theta) \\ \theta = a\xi - \omega \tau + \delta \end{array} \right.$
$u(\xi, \tau) = \dfrac{a^2}{4} sech^2 \dfrac{a}{2}[\xi - (1 + a^2) \tau + \dfrac{\delta}{a}]$

$2.非线性薛定谔方程$
$i\Phi_{tt} + \Phi_{xx} + \beta \Phi \bar \Phi^2 = 0 \quad (25)$
$令\left \lbrace \begin{array}{l} \Phi = e^{i(kx - vt)} u(\theta) \\ \theta = x-bt \end{array} \right.$
$\varphi(x, t) = a \sqrt{\dfrac{2}{\beta}} exp\lbrace i[\dfrac{1}{2}bx - (\dfrac{1}{4}b^2 - a^2)t]\rbrace sech[a(x-bt)]$

$3.正弦Gordon方程$
$\Phi_{xx} - \Phi_{tt} = \sin \Phi \quad (6)$
$\dfrac{1}{2}(\Phi + \bar \Phi)_{\xi} = \dfrac{1}{\alpha} \sin \dfrac{\Phi - \bar \Phi}{2}$
$\dfrac{1}{2}(\Phi - \bar \Phi)_{\tau} = \alpha \sin \dfrac{\Phi + \bar \Phi}{2}$
$\Phi = 4 \tan^{-1} exp[a(x-bt) + \delta]$
$a = \dfrac{\alpha^2 + 1}{2 \alpha}, b = \dfrac{1 - \alpha^2}{1 + \alpha^2}$