线性代数 01.04 克拉默法则

$\color{blue}{第一章 第四节 克拉默法则}$

$\color{blue}{一、非齐次线性方程组的克拉默法则}$

$设非其次线性方程组,(b_1, b_2, \cdots, b_n,不全为0)$
$\left \{ \begin{array}{l} a_{11}x_1 + a_{12}x_2 + \cdots + a_{1n}x_n = b_1 \\ a_{21}x_1 + a_{22}x_2 + \cdots + a_{2n}x_n = b_2 \\ \cdots \quad \cdots \quad \cdots \quad \cdots \\ a_{n1}x_1 + a_{n2}x_2 + \cdots + a_{nn}a_n = b_n \end{array} \right. \tag 1$
$若(1)的系数行列式$
$D = \begin{vmatrix} a_{11} & a_{12} & \cdots & a_{1n} \\ a_{21} & a_{22} & \cdots & a_{2n} \\ \cdots & \cdots & \cdots & \cdots \\ a_{n1} & a_{n2} & \cdots & a_{nn} \\ \end{vmatrix} \neq 0 \tag 2$
$则线性方程组(1)有唯一解$
$x_j = \dfrac{D_j}{D}, (j = 1, 2, \cdots, n) \tag 3$

$证明：先证x_j = \dfrac{D_j}{D},j = 1, 2, \cdots, n是(1)的解,\\ 要证x_j = \dfrac{D_j}{D},j = 1, 2, \cdots, n是(1)的解,只需证明\\ (3)满足(1)即可,为此把(1)改写成:\\ a_{i1}x_1 + a_{i2}x_2 + \cdots + a_{in}x_n = b_i, i = 1, 2, \cdots, n. \\ 即证明：\\ a_{i1}\dfrac{D_1}{D} + a_{i2}\dfrac{D_2}{D} + \cdots + a_{in}\dfrac{D_n}{D} = b_i, i = 1, 2, \cdots, n.$
做n + 1阶行列式
$D_{n + 1} = \begin{vmatrix} b_i & a_{i1} & \cdots & a_{ij} & \cdots & a_{in} \\ b_1 & a_{11} & \cdots & a_{1j} & \cdots & a_{1n} \\ \cdots & \cdots & \cdots & \cdots & \cdots & \cdots \\ b_i & a_{i1} & \cdots & a_{ij} & \cdots & a_{in} \\ \cdots & \cdots & \cdots & \cdots & \cdots & \cdots \\ b_n & a_{n1} & \cdots & a_{nj} & \cdots & a_{nn} \\ \end{vmatrix}$
$显然D_{n + 1} = 0.把D_{n + 1} 按第一行展开.需要求出第一行\\ 每个元素的代数余子式.第一行元素a_{ij}的代数余子式为:$
$A_{ij} = (-1)^{1 + j + 1} \begin{vmatrix} b_1 & a_{11} & \cdots & a_{1(j-1)} & a_{1(j+1)} & \cdots & a_{1n} \\ \vdots & \vdots & & \vdots & \vdots & & \vdots \\ b_n & a_{n1} & \cdots & a_{n(j-1)} & a_{n(j+1)} & \cdots & a_{nn} \\ \end{vmatrix} \\ = (-1)^{1 + j + 1}(-1)^{j-1} \begin{vmatrix} a_{11} & \cdots & a_{1(j-1)} & b_1 & a_{1(j+1)} & \cdots & a_{1n} \\ \vdots & & \vdots & \vdots & \vdots & & \vdots \\ a_{n1} & \cdots & a_{n(j-1)} & b_n & a_{n(j+1)} & \cdots & a_{nn} \\ \end{vmatrix} \\ =(-1)^{2j + 1}D_j \\ = -D_j(j = 1, 2, \cdots, n).$
$所以, D_{n + 1} = b_iD - a_{i1}D_1 - a_{i2}D_2 - \cdots - a_{in}D_n = 0 \\ 即:a_{i1}\dfrac{D_1}{D} + a_{i2}\dfrac{D_2}{D} + \cdots + a_{in}\dfrac{D_n}{D} = b_i (i = 1, 2, \cdots, n) \\ 这说明x_j = \dfrac{D_j}{D}, j = 1, 2, \cdots, n是(1)的解$

$再证唯一性,假设x_j = c_j, j = 1, 2, \cdots, n\\ 也是(1)的解,在(2)两端同时乘以c_j$
$c_jD = \begin{vmatrix} a_{11} & \cdots & a_{1j}c_j & \cdots & a_{1n} \\ \vdots & & \vdots & & \vdots \\ a_{n1} & \cdots & a_{nj}c_j & \cdots & a_{nn} \\ \end{vmatrix} \\ = \begin{vmatrix} a_{11} & \cdots & (a_{11}c_1 + \cdots + a_{1j}c_j + \cdots + a_{1n}c_n) & \cdots & a_{1n} \\ \vdots & & \vdots & & \vdots \\ a_{n1} & \cdots & (a_{n1}c_1 + \cdots + a_{nj}c_j + \cdots + a_{nn}c_n) & \cdots & a_{nn} \\ \end{vmatrix} \\ = \begin{vmatrix} a_{11} & \cdots & b_1 & \cdots & a_{1n} \\ \vdots & & \vdots & & \vdots \\ a_{n1} & \cdots & b_n & \cdots & a_{nn} \\ \end{vmatrix} \\ = D_j$
$由于 D \neq 0,所以$
$c_j = \dfrac{D_j}{D} j = 1, 2, \cdots, n.\\ 故线性方程组(1)有唯一解(3).$

例1.解线性方程组
$\left \{ \begin{array}{l} 2x_1 + x_2 -5x_3 + x_4 = 8 \\ x_1 -3x_2 \qquad \ -6x_4 = 9 \\ \qquad \ 2x_2 - x_3 + 2x_4 = -5 \\ x_1 + 4x_2 - 7x_3 + 6x_4 = 0 \end{array} \right.$
$解：\\ D = \begin{vmatrix} 2 & 1 & -5 & 1 \\ 1 & -3 & 0 & -6 \\ 0 & 2 & -1 & 2 \\ 1 & 4 & -7 & 6 \\ \end{vmatrix} \\ = \begin{vmatrix} 0 & 7 & -5 & 13 \\ 1 & -3 & 0 & -6 \\ 0 & 2 & -1 & 2 \\ 0 & 7 & -7 & 12 \\ \end{vmatrix} \\ = -\begin{vmatrix} 7 & -5 & 13 \\ 2 & -1 & 2 \\ 7 & -7 & 12 \\ \end{vmatrix} \\ = \begin{vmatrix} -3 & 3 \\ -7 & -2 \\ \end{vmatrix} = 27 \\ D_1 = \begin{vmatrix} 8 & 1 & -5 & 1 \\ 9 & -3 & 0 & -6 \\ -5 & 2 & -1 & 2 \\ 0 & 4 & -7 & 6 \\ \end{vmatrix} = 81 \\ D_2 = \begin{vmatrix} 2 & 8 & -5 & 1 \\ 1 & 9 & 0 & -6 \\ 0 & -5 & -1 & 2 \\ 1 & 0 & -7 & 6 \\ \end{vmatrix} = -108 \\ D_3 = \begin{vmatrix} 2 & 1 & 8 & 1 \\ 1 & -3 & 9 & -6 \\ 0 & 2 & -5 & 2 \\ 1 & 4 & 0 & 6 \\ \end{vmatrix} = -27 \\ D_4 = \begin{vmatrix} 2 & 1 & -5 & 8 \\ 1 & -3 & 0 & 9 \\ 0 & 2 & -1 & -5 \\ 1 & 4 & -7 & 0 \\ \end{vmatrix} = 27$
$于是得原方程组的解 \left \{ \begin{array}{l}x_1 = \dfrac{D_1}{D} = \dfrac{81}{27} = 3 \\ x_2 = \dfrac{D_2}{D} = \dfrac{-108}{27} = -4 \\ x_3 = \dfrac{D_3}{D} = \dfrac{-27}{27} = -1 \\ x_4 = \dfrac{D_4}{D} = \dfrac{27}{27} = 1 \end{array} \right.$

$定理2.如果线性方程组(1)的系数行列式D不等于0,则(1)有唯一的解.$
$定理2^{\prime}.如果线性方程组(1)的无解或有多个解,则它的系数行列式必为0.$

$\color{blue}{二、齐次线性方程组的克拉默法则}$

$设齐次线性方程组,(b_1, b_2, \cdots, b_n,不全为0)$
$\left \{ \begin{array}{l} a_{11}x_1 + a_{12}x_2 + \cdots + a_{1n}x_n = 0 \\ a_{21}x_1 + a_{22}x_2 + \cdots + a_{2n}x_n = 0 \\ \cdots \quad \cdots \quad \cdots \quad \cdots \\ a_{n1}x_1 + a_{n2}x_2 + \cdots + a_{nn}a_n = 0 \end{array} \right. \tag 4$

$若(4)的系数行列式$
$D = \begin{vmatrix} a_{11} & a_{12} & \cdots & a_{1n} \\ a_{21} & a_{22} & \cdots & a_{2n} \\ \cdots & \cdots & \cdots & \cdots \\ a_{n1} & a_{n2} & \cdots & a_{nn} \\ \end{vmatrix} \neq 0 \tag 5$
$则线性方程组(4)没有非零解$

$定理3.如果齐次线性方程组(4)的系数行列式D不等于0，则齐次线性方程组(4)没有非零解.$
$定理3^{\prime}.如果齐次线性方程组(4)有非零解,则它的系数行列式必为0.$

$例2.问\lambda在什么条件下,方程组\left \{ \begin{array}{l} \lambda x_1 + x_2 = 0 \\ x_1 + \lambda x_2 = 0 \end{array} \right. 有非零解？$
$解: 由定理3^{\prime}知，若方程组有非零解,则其系数行列式必为零\\ D = \begin{vmatrix} \lambda & 1 \\ 1 & \lambda \\ \end{vmatrix} = \lambda^2 - 1 = 0 \\ \lambda = \pm 1$

$例3.设非齐次线性方程组$
$\left \{ \begin{array}{l} \lambda x_1 + x_2 + x_3 = 1 \\ x_1 + \lambda x_2 + x_3 = \lambda \\ x_1 + x_2 + \lambda x_3 = \lambda^2 \end{array} \right.$
$问\lambda为何值时,该方程组有唯一解,并求其解.$
$解:方程组系数行列式为\\ D= \begin{vmatrix} \lambda & 1 & 1 \\ 1 & \lambda & 1 \\ 1 & 1 & \lambda \\ \end{vmatrix} = (\lambda + 2)(\lambda - 1)^2 \\ 当\lambda \neq -2, \lambda \neq 1时，方程组有唯一解。$
$D_1= \begin{vmatrix} 1 & 1 & 1 \\ \lambda & \lambda & 1 \\ \lambda^2 & 1 & \lambda \\ \end{vmatrix} = \begin{vmatrix} 0 & 1 & 1 \\ 0 & \lambda & 1 \\ \lambda^2 - 1 & 1 & \lambda \\ \end{vmatrix} \\ = (\lambda^2 - 1) \begin{vmatrix} 1 & 1 \\ \lambda & 1 \\ \end{vmatrix} = -(\lambda + 1)(\lambda - 1)^2$
$D_2= \begin{vmatrix} \lambda & 1 & 1 \\ 1 & \lambda & 1 \\ 1 & \lambda^2 & \lambda \\ \end{vmatrix} = \begin{vmatrix} 0 & 1-\lambda^2 & 1- \lambda \\ 1 & \lambda & 1 \\ 0 & \lambda^2 -\lambda & \lambda -1\\ \end{vmatrix} \\ = - \begin{vmatrix} 1-\lambda^2 & 1- \lambda \\ \lambda^2 -\lambda & \lambda -1\\ \end{vmatrix} \\ = -(1 - \lambda^2)(\lambda - 1) + (\lambda^2 - \lambda)(1 - \lambda) \\ = (\lambda - 1)^2$

$D_3= \begin{vmatrix} \lambda & 1 & 1 \\ 1 & \lambda & \lambda \\ 1 & 1 & \lambda^2 \\ \end{vmatrix} = \begin{vmatrix} \lambda & 1 & 1 \\ 1-\lambda^2 & 0 & 0 \\ 1 & 1 & \lambda^2 \\ \end{vmatrix} \\ = - (1-\lambda^2) \begin{vmatrix} 1 & 1 \\ 1 & \lambda^2 \\ \end{vmatrix} \\ = - (1-\lambda^2)(\lambda^2 - 1) \\ = (\lambda + 1)^2(\lambda - 1)^2$
$\left \{ \begin{array}{l} x_1 = \dfrac{D_1}{D} = \dfrac{-(\lambda + 1)(\lambda - 1)^2}{(\lambda + 2)(\lambda - 1)^2} = -\dfrac{\lambda + 1}{\lambda + 2} \\ x_2 = \dfrac{D_2}{D} = \dfrac{- (\lambda - 1)^2}{(\lambda + 2)(\lambda - 1)^2} = \dfrac{1}{\lambda + 2} \\ x_3 = \dfrac{D_3}{D} = \dfrac{ (\lambda + 1)^2(\lambda - 1)^2}{(\lambda + 2)(\lambda - 1)^2} = \dfrac{(\lambda + 1)^2}{\lambda + 2} \end{array} \right.$