线性代数 02.04 分块矩阵

$\color{blue}{\S 第二章 第四节 分块矩阵}$

$\color{blue}{一、分块矩阵的定义}$

把一个阶数较高的矩阵,用若干条横线和竖线分成若干小块,每一小块都叫做矩阵的子块,以子块为元素的矩阵称为分块矩阵.

$例如:将3 \times 4矩阵$
$A = \begin{pmatrix} a_{11} & a_{12} & a_{13} & a_{14} \\ a_{21} & a_{22} & a_{23} & a_{24} \\ a_{31} & a_{32} & a_{33} & a_{34} \\ \end{pmatrix}$
$分块形式如下$
(1) $\begin{pmatrix} a_{11} & a_{12} & | & a_{13} & a_{14} \\ a_{21} & a_{22} & | & a_{23} & a_{24} \\ -- & -- & | & --& -- \\ a_{31} & a_{32} & | & a_{33} & a_{34} \\ \end{pmatrix} = \begin{pmatrix} A_{11} & A_{12} \\ A_{21} & A_{22} \\ \end{pmatrix} \\ A_{11} = \begin{pmatrix} a_{11} & a_{12} \\ a_{21} & a_{22} \\ \end{pmatrix} \\ A_{12} = \begin{pmatrix} a_{13} & a_{14} \\ a_{23} & a_{24} \\ \end{pmatrix} \\ A_{21} = \begin{pmatrix} a_{31} & a_{32} \\ \end{pmatrix} \\ A_{22} = \begin{pmatrix} a_{33} & a_{34} \\ \end{pmatrix}$

(2) $\begin{pmatrix} a_{11} & | & a_{12} & a_{13} & | & a_{14} \\ a_{21} & | & a_{22} & a_{23} & | & a_{24} \\ -- & | & -- & --& | & -- \\ a_{31} & | & a_{32} & a_{33} & | & a_{34} \\ \end{pmatrix} = \begin{pmatrix} A_{11} & A_{12} & A_{13}\\ A_{21} & A_{22} & A_{23} \\ \end{pmatrix}$

(3) 按列分块
$\begin{pmatrix} a_{11} & | & a_{12} & | & a_{13} & | & a_{14} \\ a_{21} & | & a_{22} & | & a_{23} & | & a_{24} \\ a_{31} & | & a_{32} & | & a_{33} & | & a_{34} \\ \end{pmatrix} = \begin{pmatrix} A_{11} & A_{12} & A_{13} & A_{14}\\ \end{pmatrix}$

(4)
$\begin{pmatrix} a_{11} & a_{12} & a_{13} & a_{14} \\ -- & -- & --& -- \\ a_{21} & a_{22} & a_{23} & a_{24} \\ -- & -- & --& -- \\ a_{31} & a_{32} & a_{33} & a_{34} \\ \end{pmatrix} = \begin{pmatrix} A_{11} \\ A_{21} \\ A_{31} \\ \end{pmatrix}$

$\color{blue}{二、分块矩阵的运算}$

$\color{blue}{1.分块矩阵的加法}$

同型矩阵,分法相同,对应的子块相加.
$设A和B均m \times n矩阵,分法如下:\\ A = \begin{pmatrix} A_{11} & \cdots & A_{1r} \\ \vdots & & \vdots \\ A_{s1} & \cdots & A_{sr} \\ \end{pmatrix} B = \begin{pmatrix} B_{11} & \cdots & B_{1r} \\ \vdots & & \vdots \\ B_{s1} & \cdots & B_{sr} \\ \end{pmatrix} \\ 其中A_{ij} 与B_{ij}的行数相同,列数相同,那么\\ A + B = \begin{pmatrix} A_{11} + B_{11} & \cdots & A_{1r} + B_{1r} \\ \vdots & & \vdots \\ A_{s1} + B_{s1} & \cdots & A_{sr} + B_{sr} \\ \end{pmatrix} \\ 其运算律与矩阵的加法相同.$

$\color{blue}{2.分块矩阵的数乘}$

设分块矩阵
$A = \begin{pmatrix} A_{11} & \cdots & A_{1r} \\ \vdots & & \vdots \\ A_{s1} & \cdots & A_{sr} \\ \end{pmatrix}, \lambda为数,那么\\ \lambda A = \begin{pmatrix} \lambda A_{11} & \cdots & \lambda A_{1r} \\ \vdots & & \vdots \\ \lambda A_{s1} & \cdots & \lambda A_{sr} \\ \end{pmatrix} \\ 其运算律与数乘矩阵相同.$

$\color{blue}{3.分块矩阵的乘法}$

$设A为m \times l矩阵,B为l \times n矩阵,分块成\\ A = \begin{pmatrix} A_{11} & A_{12} & \cdots & A_{1t} \\ \cdots & \cdots & \cdots & \cdots \\ A_{i1} & A_{i2} & \cdots & A_{it} \\ \cdots & \cdots & \cdots & \cdots \\ A_{s1} & A_{s2} & \cdots & A_{st} \\ \end{pmatrix} B = \begin{pmatrix} B_{11} & B_{1j} & \cdots & B_{1r} \\ \cdots & \cdots & \cdots & \cdots \\ B_{i1} & B_{ij} & \cdots & B_{ir} \\ \cdots & \cdots & \cdots & \cdots \\ B_{t1} & B_{tj} & \cdots & B_{tr} \\ \end{pmatrix} \\ 其中A_{i1}, A_{i2}, \cdots, A_{it}列数分别等于\\ B_{1j}, B_{2j}, \cdots, B_{tj}的行数,那么\\ AB = \begin{pmatrix} C_{11} & C_{12} & \cdots & C_1r \\ C_{21} & C_{22} & \cdots & C_2r \\ \cdots & \cdots & \cdots & \cdots \\ C_{s1} & C_{s2} & \cdots & C_sr \\ \end{pmatrix} \\ 其中,C_{ij} = A_{i1}B_{1j} + A_{i2}B_{2j} + \cdots + A_{it}B_{tj} \\ (i = 1, 2, \cdots, s; j = 1, 2, \cdots, r)$

$例1.设 A = \begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ -1 & 2 & 1 & 0 \\ 1 & 1 & 0 & 1 \\ \end{pmatrix}, B = \begin{pmatrix} 1 & 0 & 1 & 0 \\ -1 & 2 & 0 & 1 \\ 1 & 0 & 4 & 1 \\ -1 & -1 & 2 & 0 \\ \end{pmatrix},求A + B$
$解：如下分块\\ A = \begin{pmatrix} 1 & 0 & | & 0 & 0 \\ 0 & 1 & | & 0 & 0 \\ -- & -- & -- & -- & -- \\ -1 & 2 & | & 1 & 0 \\ 1 & 1 & | & 0 & 1 \\ \end{pmatrix} = \begin{pmatrix} E & 0 \\ A_{21} & E \\ \end{pmatrix} \\ B = \begin{pmatrix} 1 & 0 & | & 1 & 0 \\ -1 & 2 & | & 0 & 1 \\ -- & -- & -- & -- & -- \\ 1 & 0 & | & 4 & 1 \\ -1 & -1 & | & 2 & 0 \\ \end{pmatrix} = \begin{pmatrix} B_{11} & E \\ B_{21} & B_{22} \\ \end{pmatrix}$
$A + B = \begin{pmatrix} E + B_{11} & 0 + E \\ A_{21} + B_{21} & E + B_{22} \\ \end{pmatrix} = \begin{pmatrix} 2 & 0 & 1 & 0 \\ -1 & 3 & 0 & 1 \\ 0 & 2 & 5 & 1 \\ 0 & 0 & 2 & 1 \\ \end{pmatrix}$

$例1.设 A = \begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ -1 & 2 & 1 & 0 \\ 1 & 1 & 0 & 1 \\ \end{pmatrix}, B = \begin{pmatrix} 1 & 0 & 1 & 0 \\ -1 & 2 & 0 & 1 \\ 1 & 0 & 4 & 1 \\ -1 & -1 & 2 & 0 \\ \end{pmatrix},求AB.$
$解:把A、B分块成\\ A = \begin{pmatrix} 1 & 0 & | & 0 & 0 \\ 0 & 1 & | & 0 & 0 \\ -- & -- & -- & -- & -- \\ -1 & 2 & | & 1 & 0 \\ 1 & 1 & | & 0 & 1 \\ \end{pmatrix} = \begin{pmatrix} E & 0 \\ A_{21} & E\\ \end{pmatrix} \\ B = \begin{pmatrix} 1 & 0 & | & 1 & 0 \\ -1 & 2 & | & 0 & 1 \\ -- & -- & -- & -- & -- \\ 1 & 0 & | & 4 & 1 \\ -1 & -1 & | & 2 & 0 \\ \end{pmatrix} = \begin{pmatrix} B_{11} & E \\ B_{21} & B_{22} \\ \end{pmatrix} \\ AB = \begin{pmatrix} E & 0 \\ A_{21} & E\\ \end{pmatrix} \begin{pmatrix} B_{11} & E \\ B_{21} & B_{22} \\ \end{pmatrix} \\ = \begin{pmatrix} B_{11} & E \\ A_{21}B_{11} + B_{21} & A_{21} + B_{22} \\ \end{pmatrix} \\ 其中\\ A_ {21}B_ {11} + B{21} = \begin{pmatrix} -1 & 2 \\ 1 & 1 \\ \end{pmatrix} \begin{pmatrix} 1 & 0 \\ -1& 2 \\ \end{pmatrix} + \begin{pmatrix} 1 & 0 \\ -1& -1 \\ \end{pmatrix} \\ = \begin{pmatrix} -3 & 4 \\ 0 & 2 \\ \end{pmatrix} + \begin{pmatrix} 1 & 0 \\ -1& -1 \\ \end{pmatrix} = \begin{pmatrix} -2 & 4 \\ -1 & 1 \\ \end{pmatrix} \\ A_{21} + B_{22} = \begin{pmatrix} -1 & 2 \\ 1 & 1 \\ \end{pmatrix} + \begin{pmatrix} 4 & 1 \\ 2 & 0 \\ \end{pmatrix} = \begin{pmatrix} 3 & 3 \\ 3 & 1 \\ \end{pmatrix} \\ 于是 AB= \begin{pmatrix} 1 & 0 & 1 & 0 \\ -1 & 2 & 0 & 1 \\ -2 & 4 & 3 & 3 \\ -1 & 1 & 3 & 1 \\ \end{pmatrix}$

$\color{blue}{4.分块矩阵的转置}$

设分块矩阵
$A = \begin{pmatrix} A_{11} & \cdots & A_{1r} \\ \vdots & & \vdots \\ A_{s1} & \cdots & A_{sr} \\ \end{pmatrix} 则 A^T = \begin{pmatrix} A_{11}^T & \cdots & A_{s1}^T \\ \vdots & & \vdots \\ A_{1r}^T & \cdots & A_{sr}^T \\ \end{pmatrix}$

$\color{blue}{5.分块对角矩阵(准对角矩阵)}$

设
$A = \begin{pmatrix} A_1 & & & \\ & A_2 & & \\ & & \ddots & \\ & & & A_s \\ \end{pmatrix} \\ 其中A_i (i = 1, 2, \cdots, s)都是方子块. \\ 显然\\ |A| = |A_1||A_2| \cdots |A_s|.$
$若|A_i| \neq 0 (i = 1, 2, \cdots, s), 则|A| \neq 0,所以$
$A^{-1} = \begin{pmatrix} A_1^{-1} & & & \\ & A_2^{-1} & & \\ & & \ddots & \\ & & & A_s^{-1} \\ \end{pmatrix}$

$例2.设A = \begin{pmatrix} 5 & 0 & 0 \\ 0 & 3 & 1 \\ 0 & 2 & 1 \\ \end{pmatrix},求A^{-1}$

$解:分块，得准对角矩阵 \begin{pmatrix} 5 & |& 0 & 0 \\ -- & -- & -- & -- \\ 0 & |& 3 & 1 \\ 0 & |& 2 & 1 \\ \end{pmatrix} = \begin{pmatrix} A_1 & \\ & A_2 \\ \end{pmatrix}\\ A_1 = \begin{pmatrix} 5 \end{pmatrix} \\ A_2 = \begin{pmatrix} 3 & 1 \\ 2 & 1 \\ \end{pmatrix} \\ |A| = |A_1||A_2| = 5 \times 1 = 5 \neq 0,有逆.\\ A_1^{-1} = \dfrac{1}{5} \\ A_2^{-1} = A_2 = \begin{pmatrix} 1 & -1 \\ -2 & 3 \\ \end{pmatrix} \\ A^{-1} = \begin{pmatrix} \frac{1}{5} & 0 & 0 \\ 0 & 1 & -1 \\ 0 & -2 & 3 \\ \end{pmatrix}$

$例3.设A的伴随矩阵\\ A^{*} = \begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & 1 & 8 \\ \end{pmatrix}\\ 且ABA^{-1} = BA^{-1} + 3E,求矩阵B.$
$解: 由|A^{*}| = |A|^{n - 1}|,有|A|^3 = 8,得|A| = 2.\\ 在ABA^{-1} = BA^{-1} + 3E的两边左乘A^{*},右乘A得\\ 2B = A^{*}B + 6E \\ (2E - A^{*})B = 6E \\ 故 B = 6(2E - A^{*})^{-1} \\ 由于 2E - A^{*} = \begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & -1 & -6 \\ \end{pmatrix} \\ (2E - A^{*})^{-1} = \begin{pmatrix} 1 & 0 & 0 & 0 \\ 0 & 1 & 0 & 0 \\ 0 & 0 & 1 & 0 \\ 0 & 0 & -\frac{1}{6} & -\frac{1}{6} \\ \end{pmatrix} B = \begin{pmatrix} 6 & 0 & 0 & 0 \\ 0 & 6 & 0 & 0 \\ 0 & 0 & 6 & 0 \\ 0 & 0 & -1 & -1 \\ \end{pmatrix}$

$\color{blue}{6.分块矩阵的应用}$

$设A为m \times n矩阵,将A按行分块,得\\ A = \begin{pmatrix} \alpha_1 \\ \alpha_2 \\ \vdots \\ \alpha_m \\ \end{pmatrix} \\ 其中 \alpha_i (i = 1, 2, \cdots, m)是A的第i行.\\ 将A按列分块,得\\ A = \begin{pmatrix} \beta_1 & \beta_2 & \cdots & \beta_n \\ \end{pmatrix}\\ 其中 \beta_j(j = 1, 2, \cdots, n)是A的第j列.$

$记A = (a_{ij})_{m \times n}, X = \begin{pmatrix} x_1 \\ x_2 \\ \vdots \\ x_n \\ \end{pmatrix}, b = \begin{pmatrix} b_1 \\ b_2 \\ \vdots \\ b_n \\ \end{pmatrix}, \\ B = \begin{pmatrix} a_{11} & a_{12} & \cdots & a_{1n} & b_1 \\ a_{21} & a_{22} & \cdots & a_{2n} & b_2 \\ \vdots & \vdots & & \vdots & \vdots \\ a_{m1} & a_{m2} & \cdots & a_{mn} & b_m \\ \end{pmatrix} \\ 其中A称为系数矩阵, x称为未知向量, b称为常数项向量,\\ B称为增广矩阵,记为: B = (A | b),\\ 或B = (A,b) = (\beta_1, \beta_2, \cdots, \beta_n, b) \\ 利用矩阵的乘法,此方程可记为： Ax = b \\ 按行分块矩阵,Ax = b又可写成：\\ \begin{pmatrix} \alpha_1^T \\ \alpha_2^T \\ \vdots \\ \alpha_m^T \\ \end{pmatrix}x = \begin{pmatrix} b_1 \\ b_2 \\ \vdots \\ b_m \\ \end{pmatrix} \\ 即a_i^T x = b_i(i = 1, 2, \cdots, m). \\ 按列分块矩阵,Ax = b又写成: \\ \begin{pmatrix} \beta_1 & \beta_2 & \cdots & \beta_n \\ \end{pmatrix} \begin{pmatrix} x_1 \\ x_2 \\ \vdots \\ x_n \\ \end{pmatrix} = b,\\ 即x_1\beta_1 + x_2\beta_2 + \cdots + x_n\beta_n = b$