返回

设有n个未知数m个方程的线性方程组(3) { a 11 x 1 + a 12 x 2 + + a 1 n x n = b 1 a 21 x 1 + a 22 x 2 + + a 2 n x n = b 2 a m 1 x 1 + a m 2 x 2 + + a m n x n = b m \left\{ \begin{array}{r} \begin{matrix} 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} \end{matrix} \\ \begin{matrix} \cdots\cdots\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \cdots\cdots \\ a_{m1}x_{1} + a_{m2}x_{2} + \cdots + a_{mn}x_{n} = b_{m} \end{matrix} \end{array} \right.\

线性方程组(3)可以写成以向量x为未知元的向量方程 Ax=b

第二章中已经说明 , 线性方程组(3)与向量方程Ax=b将混同使用而不加区分

解与解向量的名称亦不加区别

线性方程组(3)如果有解 , 就称它是相容的 ; 如果无解 , 就称它不相容

利用系数矩阵A和增广矩阵B=(A , b)的秩

可以方便地讨论线性方程组是否有解(即是否相容)

以及有解时解是否惟一等问题

定理3

(i)R(A)<R(A , b),是n元线性方程组Ax=b无解的充分必要条件。

(ii)R(A)=R(A , b)=n,是n元线性方程组Ax=b有惟一解的充分必要条件

(iii)R(A)=R(A , b)<n,是n元线性方程组Ax=b有无限多解的充分必要条件

证: 只需证明条件的充分性 ,

因为(i),(ii),(iii)中条件的必要性依次是

(ii)(iii) , (i)(iii) , (i)(ii)中条件的充分性的逆否命题

设R(A)=r . 为叙述方便 , 设B=(A , b)的行最简形矩阵为

B ̃ = [ 1 0 0 0 0 0 0 1 0 0 0 0 0 0 1 0 0 0 b 11 b 21 b r 1 0 0 0 b 1 , n r b 2 , n r b r , n r 0 0 0 d 1 d 2 d r d r + 1 0 0 ] \widetilde{B} = \left\lbrack \begin{array}{r} \begin{array}{r} \begin{matrix} 1 \\ 0 \\ \vdots \end{matrix} \\ \begin{matrix} 0 \\ 0 \\ 0 \end{matrix} \end{array} \\ \begin{matrix} \vdots \\ 0 \end{matrix} \end{array}\ \ \ \ \ \ \ \ \ \ \begin{array}{r} \begin{array}{r} \begin{matrix} 0 \\ 1 \\ \vdots \end{matrix} \\ \begin{matrix} 0 \\ 0 \\ 0 \end{matrix} \end{array} \\ \begin{matrix} \vdots \\ 0 \end{matrix} \end{array}\ \ \ \ \ \ \ \ \ \ \begin{array}{r} \begin{array}{r} \begin{matrix} \cdots \\ \cdots \\ \end{matrix} \\ \begin{matrix} \cdots \\ \cdots \\ \cdots \end{matrix} \end{array} \\ \begin{matrix} \\ \cdots \end{matrix} \end{array}\ \ \ \ \ \ \ \ \ \ \begin{array}{r} \begin{array}{r} \begin{matrix} 0 \\ 0 \\ \vdots \end{matrix} \\ \begin{matrix} 1 \\ 0 \\ 0 \end{matrix} \end{array} \\ \begin{matrix} \vdots \\ 0 \end{matrix} \end{array}\ \ \ \ \ \ \ \ \ \ \begin{array}{r} \begin{array}{r} \begin{matrix} b_{11} \\ b_{21} \\ \vdots \end{matrix} \\ \begin{matrix} b_{r1} \\ 0 \\ 0 \end{matrix} \end{array} \\ \begin{matrix} \vdots \\ 0 \end{matrix} \end{array}\ \ \ \ \ \ \ \ \ \ \begin{array}{r} \begin{array}{r} \begin{matrix} \cdots \\ \cdots \\ \end{matrix} \\ \begin{matrix} \cdots \\ \cdots \\ \cdots \end{matrix} \end{array} \\ \begin{matrix} \\ \cdots \end{matrix} \end{array}\ \ \ \ \ \ \ \ \ \ \begin{array}{r} \begin{array}{r} \begin{matrix} b_{1\ ,\ n - r} \\ b_{2\ ,\ n - r} \\ \vdots \end{matrix} \\ \begin{matrix} b_{r\ ,\ n - r} \\ 0 \\ 0 \end{matrix} \end{array} \\ \begin{matrix} \vdots \\ 0 \end{matrix} \end{array}\ \ \ \ \ \ \ \ \ \ \begin{array}{r} \begin{array}{r} \begin{matrix} d_{1} \\ d_{2} \\ \vdots \end{matrix} \\ \begin{matrix} d_{r} \\ d_{r + 1} \\ 0 \end{matrix} \end{array} \\ \begin{matrix} \vdots \\ 0 \end{matrix} \end{array}\ \ \right\rbrack

(i)若R(A)<R(B) , 则 B ̃ \widetilde{B} 中的dr+1=1 , 于是 B ̃ \widetilde{B} 的第r+1行对应矛盾方程0=1

故方程(4)无解

(ii)若R(A)=R(B) , 则进一步把B化成行最简形矩阵

而对于齐次线性方程组 , 则把系数矩阵A化成行最简形矩阵

(iii)设R(A)=R(B)=r

把行最简形中r个非零行的首非零元所对应的未知数取作非自由未知数

其余n-r个未知数取作自由未知数

并令自由未知数分别等于c , c2 , ⋯ , cn-r , 由B(或A)是行最简形矩阵

即可写出含n-r个参数的通解

例10求解齐次线性方程组 { x 1 + 2 x 2 + 2 x 3 + x 4 = 0 2 x 1 + x 2 2 x 3 2 x 4 = 0 x 1 x 2 4 x 3 3 x 4 = 0 \left\{ \begin{matrix} x_{1} + 2x_{2} + 2x_{3} + x_{4} = 0 \\ 2x_{1} + x_{2} - 2x_{3} - 2x_{4} = 0 \\ x_{1} - x_{2} - 4x_{3} - 3x_{4} = 0 \end{matrix} \right.\

解: 对系数矩阵A施行初等行变换变为行最简形矩阵

A= [ 1 2 1 2 1 1 2 2 4 1 2 3 ] r 2 2 r 1 r 3 r 1 [ 1 0 0 2 3 3 2 6 6 1 4 4 ] \left\lbrack \begin{matrix} 1 \\ 2 \\ 1 \end{matrix}\ \ \ \begin{matrix} 2 \\ 1 \\ - 1 \end{matrix}\ \ \ \begin{matrix} 2 \\ - 2 \\ - 4 \end{matrix}\ \ \ \begin{matrix} 1 \\ - 2 \\ - 3 \end{matrix} \right\rbrack\overset{\begin{matrix} r_{2} - 2r_{1} \\ r_{3} - r_{1} \end{matrix}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \begin{matrix} 2 \\ - 3 \\ - 3 \end{matrix}\ \ \ \begin{matrix} 2 \\ - 6 \\ - 6 \end{matrix}\ \ \ \begin{matrix} 1 \\ - 4 \\ - 4 \end{matrix} \right\rbrack

r 3 r 2 r 2 ÷ ( 3 ) [ 1 0 0 2 1 0 2 2 0 1 4 3 0 ] r 1 2 r 2 [ 1 0 0 0 1 0 2 2 0 5 3 5 3 0 ] \overset{\begin{matrix} r_{3} - r_{2} \\ r_{2} \div ( - 3) \end{matrix}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \begin{matrix} 2 \\ 1 \\ 0 \end{matrix}\ \ \ \begin{matrix} 2 \\ 2 \\ 0 \end{matrix}\ \ \ \begin{matrix} 1 \\ \frac{4}{3} \\ 0 \end{matrix} \right\rbrack\overset{r_{1} - 2r_{2}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \begin{matrix} 0 \\ 1 \\ 0 \end{matrix}\ \ \ \begin{matrix} - 2 \\ - 2 \\ 0 \end{matrix}\ \ \ \begin{matrix} \frac{- 5}{3} \\ \frac{- 5}{3} \\ 0 \end{matrix} \right\rbrack

得同解的方程组 { x 1 2 x 3 5 3 x 4 = 0 x 2 + 2 x 3 + 4 3 x 4 = 0 \left\{ \begin{matrix} x_{1} - 2x_{3} - \frac{5}{3}x_{4} = 0 \\ x_{2} + 2x_{3} + \frac{4}{3}x_{4} = 0 \end{matrix} \right.\

得参数形式 { x 1 = 2 x 3 + 5 3 x 4 x 2 = 2 x 3 4 3 x 4 \left\{ \begin{matrix} x_{1} = 2x_{3} + \frac{5}{3}x_{4} \\ x_{2} = - 2x_{3} - \frac{4}{3}x_{4} \end{matrix} \right.\ (x3 , x4可任意取值)

得参数形式 { x 1 = 2 c 1 + 5 3 c 2 x 2 = 2 c 1 4 3 c 2 x 3 = c 1 x 4 = c 2 \left\{ \begin{array}{r} \begin{matrix} x_{1} = 2c_{1} + \frac{5}{3}c_{2} \\ x_{2} = - 2c_{1} - \frac{4}{3}c_{2} \end{matrix} \\ \begin{matrix} x_{3} = c_{1}\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \\ x_{4} = c_{2}\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \end{matrix} \end{array} \right.\ 令x3=c1 , x4=c2 , 其中c1 , c2为任意实数

得齐次通解 [ x 1 x 2 x 3 x 4 ] \left\lbrack \begin{array}{r} \begin{matrix} x_{1} \\ x_{2} \end{matrix} \\ \begin{matrix} x_{3} \\ x_{4} \end{matrix} \end{array} \right\rbrack = [ 2 c 1 + 5 3 c 2 2 c 1 4 3 c 2 c 1 c 2 ] \left\lbrack \begin{array}{r} \begin{matrix} 2c_{1} + \frac{5}{3}c_{2} \\ - 2c_{1} - \frac{4}{3}c_{2} \end{matrix} \\ \begin{matrix} c_{1} \\ \ c_{2} \end{matrix} \end{array} \right\rbrack = c 1 [ 2 2 1 0 ] c_{1}\left\lbrack \begin{array}{r} \begin{matrix} 2 \\ - 2 \end{matrix} \\ \begin{matrix} 1 \\ 0 \end{matrix} \end{array} \right\rbrack + c 2 [ 5 3 4 3 0 1 ] c_{2}\left\lbrack \begin{array}{r} \begin{matrix} \frac{5}{3} \\ \frac{- 4}{3} \end{matrix} \\ \begin{matrix} 0 \\ 1 \end{matrix} \end{array} \right\rbrack

例11求解非齐次线性方程组 { x 1 2 x 2 + 3 x 3 x 4 = 1 3 x 1 x 2 + 5 x 3 3 x 4 = 2 2 x 1 + x 2 + x 3 2 x 4 = 3 \left\{ \begin{matrix} x_{1} - 2x_{2} + 3x_{3} - x_{4} = 1 \\ 3x_{1} - x_{2} + 5x_{3} - 3x_{4} = 2 \\ 2x_{1} + x_{2} + x_{3} - 2x_{4} = 3 \end{matrix} \right.\

解: 对增广矩阵B施行初等行变换

B= [ 1 3 2 2 1 1 3 5 2 1 3 2 1 2 3 ] r 2 3 r 1 r 3 2 r 1 [ 1 0 0 2 5 5 3 4 4 1 0 0 1 1 1 ] r 3 r 2 [ 1 0 0 2 5 0 3 4 0 1 0 0 1 1 2 ] \left\lbrack \begin{matrix} 1 \\ 3 \\ 2 \end{matrix}\ \ \ \begin{matrix} - 2 \\ - 1 \\ 1 \end{matrix}\ \ \ \ \begin{matrix} 3 \\ 5 \\ 2 \end{matrix}\ \ \ \begin{matrix} - 1 \\ - 3 \\ - 2 \end{matrix}\ \ \ \ \begin{matrix} 1 \\ 2 \\ 3 \end{matrix} \right\rbrack\overset{\begin{matrix} r_{2} - 3r_{1} \\ r_{3} - 2r_{1} \end{matrix}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \begin{matrix} - 2 \\ 5 \\ 5 \end{matrix}\ \ \ \begin{matrix} 3 \\ - 4 \\ - 4 \end{matrix}\ \ \ \ \begin{matrix} - 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \begin{matrix} 1 \\ - 1 \\ 1 \end{matrix} \right\rbrack\overset{r_{3} - r_{2}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \begin{matrix} - 2 \\ 5 \\ 0 \end{matrix}\ \ \ \begin{matrix} 3 \\ - 4 \\ 0 \end{matrix}\ \ \ \ \begin{matrix} - 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \begin{matrix} 1 \\ - 1 \\ 2 \end{matrix} \right\rbrack

可见R(A)=2 , R(B)=3 , 故方程组无解

例12求解非齐次线性方程组 { x 1 + x 2 3 x 3 x 4 = 1 3 x 1 x 2 3 x 3 + 4 x 4 = 4 x 1 + 5 x 2 9 x 3 8 x 4 = 0 \left\{ \begin{matrix} x_{1} + x_{2} - 3x_{3} - x_{4} = 1 \\ 3x_{1} - x_{2} - 3x_{3} + 4x_{4} = 4 \\ x_{1} + 5x_{2} - 9x_{3} - 8x_{4} = 0 \end{matrix} \right.\

解: 对增广矩阵B施行初等行变换

B= [ 1 3 1 1 1 5 3 3 9 1 4 8 1 4 0 ] r 2 3 r 1 r 3 r 1 [ 1 0 0 1 4 4 3 6 6 1 7 7 1 1 1 ] \left\lbrack \begin{matrix} 1 \\ 3 \\ 1 \end{matrix}\ \ \ \begin{matrix} 1 \\ - 1 \\ 5 \end{matrix}\ \ \ \ \begin{matrix} - 3 \\ - 3 \\ - 9 \end{matrix}\ \ \ \begin{matrix} - 1 \\ 4 \\ - 8 \end{matrix}\ \ \ \ \begin{matrix} 1 \\ 4 \\ 0 \end{matrix} \right\rbrack\overset{\begin{matrix} r_{2} - 3r_{1} \\ r_{3} - r_{1} \end{matrix}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \begin{matrix} 1 \\ - 4 \\ 4 \end{matrix}\ \ \ \begin{matrix} - 3 \\ 6 \\ - 6 \end{matrix}\ \ \ \ \begin{matrix} - 1 \\ 7 \\ - 7 \end{matrix}\ \ \ \ \begin{matrix} 1 \\ 1 \\ - 1 \end{matrix} \right\rbrack

r 3 + r 2 r 2 ÷ ( 4 ) [ 1 0 0 1 1 0 3 3 2 0 1 7 4 0 1 1 4 0 ] r 1 r 2 [ 1 0 0 0 1 0 3 2 3 2 0 3 4 7 4 0 5 4 1 4 0 ] \overset{\begin{matrix} r_{3} + r_{2} \\ r_{2} \div ( - 4) \end{matrix}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \begin{matrix} 1 \\ 1 \\ 0 \end{matrix}\ \ \ \begin{matrix} - 3 \\ \frac{- 3}{2} \\ 0 \end{matrix}\ \ \ \ \begin{matrix} - 1 \\ \frac{- 7}{4} \\ 0 \end{matrix}\ \ \ \ \begin{matrix} 1 \\ \frac{- 1}{4} \\ 0 \end{matrix} \right\rbrack\overset{r_{1} - r_{2}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \begin{matrix} 0 \\ 1 \\ 0 \end{matrix}\ \ \ \begin{matrix} \frac{- 3}{2} \\ \frac{- 3}{2} \\ 0 \end{matrix}\ \ \ \ \begin{matrix} \frac{3}{4} \\ \frac{- 7}{4} \\ 0 \end{matrix}\ \ \ \ \begin{matrix} \frac{5}{4} \\ \frac{- 1}{4} \\ 0 \end{matrix} \right\rbrack

得同解方程组 { x 1 + 0 3 2 x 3 + 3 4 x 4 = 5 4 0 + x 2 3 2 x 3 7 4 x 4 = 1 4 \left\{ \begin{matrix} x_{1} + 0 - \frac{3}{2}x_{3} + \frac{3}{4}x_{4} = \frac{5}{4} \\ 0 + x_{2} - \frac{3}{2}x_{3} - \frac{7}{4}x_{4} = - \frac{1}{4} \end{matrix} \right.\

得参数形式 { x 1 = 3 2 x 3 3 4 x 4 + 5 4 x 2 = 3 2 x 3 + 7 4 x 4 1 4 \left\{ \begin{matrix} x_{1} = \frac{3}{2}x_{3} - \frac{3}{4}x_{4} + \frac{5}{4} \\ x_{2} = \frac{3}{2}x_{3} + \frac{7}{4}x_{4} - \frac{1}{4} \end{matrix} \right.\ (x3 , x4可任意取值)

得参数形式 { x 1 = 3 2 c 1 3 4 c 2 + 5 4 x 2 = 3 2 c 1 + 7 4 c 2 1 4 x 3 = c 1 x 4 = c 2 \left\{ \begin{array}{r} \begin{matrix} x_{1} = \frac{3}{2}c_{1} - \frac{3}{4}c_{2} + \frac{5}{4} \\ x_{2} = \frac{3}{2}c_{1} + \frac{7}{4}c_{2} - \frac{1}{4} \end{matrix} \\ \begin{matrix} x_{3} = c_{1}\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \\ x_{4} = c_{2}\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \end{matrix} \end{array} \right.\

得非齐次通解 [ x 1 x 2 x 3 x 4 ] = [ 3 2 c 1 3 4 c 2 + 5 4 3 2 c 1 + 7 4 c 2 1 4 c 1 c 2 ] \left\lbrack \begin{array}{r} \begin{matrix} x_{1} \\ x_{2} \end{matrix} \\ \begin{matrix} x_{3} \\ x_{4} \end{matrix} \end{array} \right\rbrack = \left\lbrack \begin{array}{r} \begin{matrix} \frac{3}{2}c_{1} - \frac{3}{4}c_{2} + \frac{5}{4} \\ \frac{3}{2}c_{1} + \frac{7}{4}c_{2} - \frac{1}{4} \end{matrix} \\ \begin{matrix} c_{1} \\ \ c_{2} \end{matrix} \end{array} \right\rbrack = c 1 [ 3 2 3 2 1 0 ] c_{1}\left\lbrack \begin{array}{r} \begin{matrix} \frac{3}{2} \\ \frac{3}{2} \end{matrix} \\ \begin{matrix} 1 \\ 0 \end{matrix} \end{array} \right\rbrack + c 2 [ 3 4 7 4 0 1 ] c_{2}\left\lbrack \begin{array}{r} \begin{matrix} \frac{- 3}{4} \\ \frac{7}{4} \end{matrix} \\ \begin{matrix} 0 \\ 1 \end{matrix} \end{array} \right\rbrack + [ 5 4 1 4 0 0 ] \left\lbrack \begin{array}{r} \begin{matrix} \frac{5}{4} \\ \frac{- 1}{4} \end{matrix} \\ \begin{matrix} 0 \\ 0 \end{matrix} \end{array} \right\rbrack , (c1 , c2 ∈ R)

例13设有线性方程组 { ( 1 + λ ) x 1 + x 2 + x 3 = 0 x 1 + ( 1 + λ ) x 2 + x 3 = 3 x 1 + x 2 + ( 1 + λ ) x 3 = λ \left\{ \begin{matrix} (1 + \lambda)x_{1} + x_{2} + x_{3} = 0 \\ x_{1} + (1 + \lambda)x_{2} + x_{3} = 3 \\ x_{1} + x_{2} + (1 + \lambda)x_{3} = \lambda \end{matrix} \right.\

问λ取何值时 , 此方程组(1)无解 ; (2)有惟一解 ; (3)有无限多解?

并在有无限多解时求其通解

解法1对增广矩阵B=(A , b)作初等行变换把它变为行阶梯形矩阵

有B= [ 1 + λ 1 1 1 1 + λ 1 1 1 1 + λ 0 3 λ ] r 1 r 3 [ 1 1 1 + λ 1 1 + λ 1 1 + λ 1 1 λ 3 0 ] \left\lbrack \begin{matrix} 1 + \lambda \\ 1 \\ 1 \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ 1 + \lambda \\ 1 \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 \\ 1 \\ 1 + \lambda \end{matrix}\ \ \ \ \ \begin{matrix} 0 \\ 3 \\ \lambda \end{matrix}\ \right\rbrack\overset{r_{1} \leftrightarrow r_{3}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 1 \\ 1 + \lambda \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ 1 + \lambda \\ 1 \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 + \lambda \\ 1 \\ 1 \end{matrix}\ \ \ \ \ \begin{matrix} \lambda \\ 3 \\ 0 \end{matrix}\ \right\rbrack

r 2 r 1 r 3 ( 1 + λ ) r 1 [ 1 0 0 1 λ λ 1 + λ λ λ ( 2 + λ ) λ 3 λ λ ( 1 + λ ) ] \overset{\begin{matrix} r_{2} - r_{1} \\ r_{3} - (1 + \lambda)r_{1} \end{matrix}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ \lambda \\ - \lambda \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 + \lambda \\ - \lambda \\ - \lambda(2 + \lambda) \end{matrix}\ \ \ \ \ \begin{matrix} \lambda \\ 3 - \lambda \\ - \lambda(1 + \lambda) \end{matrix}\ \right\rbrack

r 3 + r 2 [ 1 0 0 1 λ 0 1 + λ λ λ ( 3 + λ ) λ 3 λ ( 1 λ ) ( 3 + λ ) ] \overset{r_{3} + r_{2}}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ \lambda \\ 0 \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 + \lambda \\ - \lambda \\ - \lambda(3 + \lambda) \end{matrix}\ \ \ \ \ \begin{matrix} \lambda \\ 3 - \lambda \\ (1 - \lambda)(3 + \lambda) \end{matrix}\ \right\rbrack

(1)当λ=0时 ,

[ 1 0 0 1 λ 0 1 + λ λ λ ( 3 + λ ) λ 3 λ ( 1 λ ) ( 3 + λ ) ] = [ 1 0 0 1 0 0 1 0 0 0 3 3 ] = [ 1 0 0 1 0 0 1 0 0 0 1 0 ] \left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ \lambda \\ 0 \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 + \lambda \\ - \lambda \\ - \lambda(3 + \lambda) \end{matrix}\ \ \ \ \ \begin{matrix} \lambda \\ 3 - \lambda \\ (1 - \lambda)(3 + \lambda) \end{matrix}\ \right\rbrack = \left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 0 \\ 3 \\ 3 \end{matrix} \right\rbrack = \left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 0 \\ 1 \\ 0 \end{matrix} \right\rbrack

R(A)=1 , R(B)=2 , 方程组无解

(2)当λ≠0且λ≠-3时 , [ 1 0 0 1 λ 0 1 + λ λ λ ( 3 + λ ) λ 3 λ ( 1 λ ) ( 3 + λ ) ] \left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ \lambda \\ 0 \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 + \lambda \\ - \lambda \\ - \lambda(3 + \lambda) \end{matrix}\ \ \ \ \ \begin{matrix} \lambda \\ 3 - \lambda \\ (1 - \lambda)(3 + \lambda) \end{matrix}\ \right\rbrack

R(A)=R(B)=3 , 方程组有惟一解

(3)当λ=-3时 ,

[ 1 0 0 1 λ 0 1 + λ λ λ ( 3 + λ ) λ 3 λ ( 1 λ ) ( 3 + λ ) ] r [ 1 0 0 1 3 0 2 3 0 3 6 0 ] r [ 1 0 0 0 1 0 1 1 0 1 2 0 ] \left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \begin{matrix} 1 \\ \lambda \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 1 + \lambda \\ - \lambda \\ - \lambda(3 + \lambda) \end{matrix}\ \ \ \ \begin{matrix} \lambda \\ 3 - \lambda \\ (1 - \lambda)(3 + \lambda) \end{matrix}\ \right\rbrack\overset{r}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \begin{matrix} 1 \\ - 3 \\ 0 \end{matrix}\ \ \ \ \begin{matrix} - 2 \\ 3 \\ 0 \end{matrix}\ \ \ \begin{matrix} - 3 \\ 6 \\ 0 \end{matrix}\ \right\rbrack\overset{r}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \begin{matrix} 0 \\ 1 \\ 0 \end{matrix}\ \ \ \ \begin{matrix} - 1 \\ - 1 \\ 0 \end{matrix}\ \ \ \begin{matrix} - 1 \\ - 2 \\ 0 \end{matrix}\ \right\rbrack

R(A)=R(B)=2 , 方程组有无限多个解

得同解方程组 { x 1 x 3 = 1 x 2 x 3 = 2 \left\{ \begin{matrix} x_{1} - x_{3} = - 1 \\ x_{2} - x_{3} = - 2 \end{matrix} \right.\

得参数形式 { x 1 = x 3 1 x 2 = x 3 2 \left\{ \begin{matrix} x_{1} = x_{3} - 1 \\ x_{2} = x_{3} - 2 \end{matrix} \right.\ (x3可任意取值)

得参数形式 { x 1 = c 1 x 2 = c 2 x 3 = c \left\{ \begin{matrix} x_{1} = c - 1 \\ x_{2} = c - 2 \\ x_{3} = c\ \ \ \ \ \ \ \end{matrix} \right.\ 令x3=c , 其中c为任意实数

得非齐次通解 [ x 1 x 2 x 3 ] \begin{bmatrix} x_{1} \\ x_{2} \\ x_{3} \end{bmatrix} =c [ 1 1 1 ] \begin{bmatrix} 1 \\ 1 \\ 1 \end{bmatrix} + [ 1 2 0 ] \begin{bmatrix} - 1 \\ - 2 \\ 0 \end{bmatrix} (c属于R)

解法2因系数矩阵A为3阶方阵 , 故有R(A)⩽R(A , b)3×4⩽3

于是由定理3 , 知

A的秩R(A)=3 , 即|A|≠0是方程有惟一解的充分必要条件。

而|A|= | 1 + λ 1 1 1 1 + λ 1 1 1 1 + λ | = ( 3 + λ ) | 1 1 1 1 1 + λ 1 1 1 1 + λ | \left| \begin{matrix} 1 + \lambda & 1 & 1 \\ 1 & 1 + \lambda & 1 \\ 1 & 1 & 1 + \lambda \end{matrix} \right| = (3 + \lambda)\left| \begin{matrix} 1 & 1 & 1 \\ 1 & 1 + \lambda & 1 \\ 1 & 1 & 1 + \lambda \end{matrix} \right|

= ( 3 + λ ) | 1 1 1 0 λ 0 1 0 λ | = (3 + \lambda)\left| \begin{matrix} 1 & 1 & 1 \\ 0 & \lambda & 0 \\ 1 & 0 & \lambda \end{matrix} \right| =(3+ λ \lambda ) λ \lambda 2

当λ=0时 , B= [ 1 1 1 1 1 1 1 1 1 0 3 0 ] r [ 1 0 0 1 0 0 1 0 0 0 1 0 ] \left\lbrack \begin{matrix} 1 \\ 1 \\ 1 \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ 1 \\ 1 \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 \\ 1 \\ 1 \end{matrix}\ \ \ \ \ \begin{matrix} 0 \\ 3 \\ 0 \end{matrix} \right\rbrack\overset{r}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 0 \\ 1 \\ 0 \end{matrix}\ \right\rbrack

知R(A)=1 , R(B)=2 , 故方程组无解

当λ≠0且λ≠-3时 , 方程组有惟一解

当λ=-3时 , B= [ 2 1 1 1 2 1 1 1 2 0 3 3 ] r [ 1 0 0 0 1 0 1 1 0 1 2 0 ] \left\lbrack \begin{matrix} - 2 \\ 1 \\ 1 \end{matrix}\ \ \ \ \ \begin{matrix} 1 \\ - 2 \\ 1 \end{matrix}\ \ \ \ \ \ \begin{matrix} 1 \\ 1 \\ - 2 \end{matrix}\ \ \ \ \ \begin{matrix} 0 \\ 3 \\ - 3 \end{matrix} \right\rbrack\overset{r}{\Rightarrow}\left\lbrack \begin{matrix} 1 \\ 0 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} 0 \\ 1 \\ 0 \end{matrix}\ \ \ \ \ \ \begin{matrix} - 1 \\ - 1 \\ 0 \end{matrix}\ \ \ \ \ \begin{matrix} - 1 \\ - 2 \\ 0 \end{matrix}\ \right\rbrack

知 R(A)=R(B)=2 , 故方程组有无限多个解

且通解为 [ x 1 x 2 x 3 ] \begin{bmatrix} x_{1} \\ x_{2} \\ x_{3} \end{bmatrix} =c [ 1 1 1 ] \begin{bmatrix} 1 \\ 1 \\ 1 \end{bmatrix} + [ 1 2 0 ] \begin{bmatrix} - 1 \\ - 2 \\ 0 \end{bmatrix} (c属于R)

比较解法1与解法2 , 显见解法2较简单

但解法2的方法只适用于系数矩阵为方阵的情形

对含参数的矩阵作初等变换时,

例如在本例中对矩阵B作初等变换时,

由于λ+1 , λ+3等因式可以等于0,

故不宜作诸如r2 - 1 λ + 1 \frac{1}{\lambda + 1} r1 , r2×(λ+1) , r3÷(λ+3)这样的变换。

如果作了这种变换,则需对λ+1=0(或λ+3=0)的情形另作讨论。

因此 , 对含参数的矩阵作初等变换较不方便

定理4 R(A)<n,是n元齐次线性方程组Ax=0有无限多解的充分必要条件

此定理是当,

R(A)=R(A , b)<n,是n元线性方程组Ax=b有无限多解的充分必要条件,

b=0时的特殊情况

定理5 R(A)=R(A , b),是非齐次线性方程组Ax=b有解的充分必要条件

定理6 R(A)=R(A , B),是矩阵方程AX=B有解的充分必要条件

证:

核心概念回顾

1.矩阵方程 A X = B AX = B 的结构:

A A m × n m \times n 矩阵, B B m × l m \times l 矩阵, X X n × l n \times l 未知矩阵。

若将 B B 按列分块为 B = ( β 1 , β 2 , , β l ) B = (\beta_{1},\beta_{2},\ldots,\beta_{l}) X X 按列分块为 X = ( x 1 , x 2 , , x l ) X = (x_{1},x_{2},\ldots,x_{l})

则矩阵方程 A X = B AX = B 等价于 l l 个线性方程组: A x 1 = β 1 , A x 2 = β 2 , , A x l = β l Ax_{1} = \beta_{1},\ Ax_{2} = \beta_{2},\ \ldots,\ Ax_{l} = \beta_{l}

2.线性方程组有解的充要条件:

R ( A ) = R ( A , β ) R(A) = R(A,\beta) 是单个线性方程组 A x = β Ax = \beta 有解的充要条件

(系数矩阵的秩等于增广矩阵的秩)。

证明过程

1.充分性(若 R ( A ) = R ( A , B ) R(A) = R(A,B) ,则 A X = B AX = B 有解)

R ( A ) = R ( A , B ) R(A) = R(A,B) ,则对每个列向量 β j \beta_{j} j = 1 , 2 , , l j = 1,2,\ldots,l ),

考虑增广矩阵 ( A , β j ) (A,\beta_{j}) 的秩:

由于 ( A , β j ) (A,\beta_{j}) ( A , B ) (A,B) 的子矩阵(仅保留第 j j 列),根据矩阵秩的性质: R ( A ) R ( A , β j ) R ( A , B ) R(A) \leq R(A,\beta_{j}) \leq R(A,B)

结合已知 R ( A ) = R ( A , B ) R(A) = R(A,B) ,得: R ( A ) = R ( A , β j ) R(A) = R(A,\beta_{j})

根据线性方程组有解的充要条件,每个 A x j = β j Ax_{j} = \beta_{j} 都有解。

因此,存在 x 1 , x 2 , , x l x_{1},x_{2},\ldots,x_{l} 使得 A X = B AX = B ,即矩阵方程 A X = B AX = B 有解。

2.必要性(若 A X = B AX = B 有解,则 R ( A ) = R ( A , B ) R(A) = R(A,B)

若矩阵方程 A X = B AX = B 有解,

则对每个列向量 β j \beta_{j} j = 1 , 2 , , l j = 1,2,\ldots,l ),线性方程组 A x j = β j Ax_{j} = \beta_{j} 都有解。

根据线性方程组有解的充要条件:

对每个 j j ,有 R ( A ) = R ( A , β j ) R(A) = R(A,\beta_{j})

由于 R ( A , B ) R(A,B) 是增广矩阵 ( A , β 1 , β 2 , , β l ) (A,\beta_{1},\beta_{2},\ldots,\beta_{l}) 的秩,而每个 R ( A , β j ) = R ( A ) R(A,\beta_{j}) = R(A)

因此: R ( A ) R ( A , B ) R ( A ) + R(A) \leq R(A,B) \leq R(A) + 常数

但更严格地, R ( A , B ) R(A,B) 的秩由 A A B B 的列向量组共同决定。

由于每个 β j \beta_{j} 可由 A A 的列向量组线性表示(因 A x j = β j Ax_{j} = \beta_{j} 有解),

B B 的列向量组可由 A A 的列向量组线性表示。

因此,增广矩阵 ( A , B ) (A,B) 的列向量组的秩与 A A 的列向量组的秩相等,

即: R ( A , B ) = R ( A ) R(A,B) = R(A)

3.综上, R ( A ) = R ( A , B ) R(A) = R(A,B) 是矩阵方程 A X = B AX = B 有解的充分必要条件

定理7 设AB=C , 则R(C)⩽min{R(A) , R(B)}

即矩阵乘积 C C 的秩不超过 A A B B 的秩的最小值,

证:

一、证明 R ( C ) R ( A ) R(C) \leq R(A)

矩阵的秩可以定义为其行向量组的秩(或列向量组的秩)。

设矩阵 A = ( a i j ) m × n A = (a_{ij})_{m \times n} B = ( b i j ) n × p B = (b_{ij})_{n \times p} ,则 C = A B = ( c i j ) m × p C = AB = (c_{ij})_{m \times p} ,其中 c i j = k = 1 n a i k b k j c_{ij} = \sum_{k = 1}^{n}a_{ik}b_{kj}

从列向量角度分析:

C C 的第 j j c j = ( c 1 j c 2 j c m j ) = k = 1 n b k j ( a 1 k a 2 k a m k ) = k = 1 n b k j a k c_{j} = \begin{pmatrix} c_{1j} \\ c_{2j} \\ \vdots \\ c_{mj} \end{pmatrix} = \sum_{k = 1}^{n}b_{kj}\begin{pmatrix} a_{1k} \\ a_{2k} \\ \vdots \\ a_{mk} \end{pmatrix} = \sum_{k = 1}^{n}b_{kj}a_{k} ,其中 a k a_{k} A A 的第 k k 列。

这表明: C C 的每一列都是 A A 的列向量组的线性组合。

根据秩的性质:若向量组 β 1 , β 2 , , β p \beta_{1},\beta_{2},\ldots,\beta_{p} 可由向量组 α 1 , α 2 , , α n \alpha_{1},\alpha_{2},\ldots,\alpha_{n} 线性表示,

r a n k ( β 1 , , β p ) r a n k ( α 1 , , α n ) rank(\beta_{1},\ldots,\beta_{p}) \leq rank(\alpha_{1},\ldots,\alpha_{n})

因此, C C 的列向量组的秩(即 R ( C ) R(C) )小于等于 A A 的列向量组的秩(即 R ( A ) R(A) ),

即: R ( C ) R ( A ) R(C) \leq R(A)

二、证明 R ( C ) R ( B ) R(C) \leq R(B)

从行向量角度分析:

C C 的第 i i c i = ( c i 1 , c i 2 , , c i p ) = k = 1 n a i k ( b k 1 , b k 2 , , b k p ) = k = 1 n a i k b k c_{i} = \begin{pmatrix} c_{i1},c_{i2},\ldots,c_{ip} \end{pmatrix} = \sum_{k = 1}^{n}a_{ik}\begin{pmatrix} b_{k1},b_{k2},\ldots,b_{kp} \end{pmatrix} = \sum_{k = 1}^{n}a_{ik}b_{k}

其中 b k b_{k} B B 的第 k k 行。

这表明: C C 的每一行都是 B B 的行向量组的线性组合。

同理,根据秩的性质:若向量组 γ 1 , γ 2 , , γ m \gamma_{1},\gamma_{2},\ldots,\gamma_{m} 可由向量组 δ 1 , δ 2 , , δ n \delta_{1},\delta_{2},\ldots,\delta_{n} 线性表示,

r a n k ( γ 1 , , γ m ) r a n k ( δ 1 , , δ n ) rank(\gamma_{1},\ldots,\gamma_{m}) \leq rank(\delta_{1},\ldots,\delta_{n})

因此, C C 的行向量组的秩(即 R ( C ) R(C) )小于等于 B B 的行向量组的秩(即 R ( B ) R(B) ),

即: R ( C ) R ( B ) R(C) \leq R(B)

三、结论由上述两部分证明可知: R ( C ) R ( A ) R(C) \leq R(A) R ( C ) R ( B ) R(C) \leq R(B)

因此, R ( C ) m i n { R ( A ) , R ( B ) } R(C) \leq min\{ R(A),R(B)\}

下面通过具体例子验证“若 A B = C AB = C ,则 R ( C ) m i n { R ( A ) , R ( B ) } R(C) \leq min\{ R(A),R(B)\} ”这一结论。

例1:满秩矩阵相乘

A = ( 1 0 0 1 ) A = \begin{pmatrix} 1 & 0 \\ 0 & 1 \end{pmatrix} (2阶单位矩阵, R ( A ) = 2 R(A) = 2 ), B = ( 1 2 3 4 ) B = \begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix} (2阶可逆矩阵, R ( B ) = 2 R(B) = 2 )。

计算乘积: C = A B = ( 1 × 1 + 0 × 3 1 × 2 + 0 × 4 0 × 1 + 1 × 3 0 × 2 + 1 × 4 ) = ( 1 2 3 4 ) C = AB = \begin{pmatrix} 1 \times 1 + 0 \times 3 & 1 \times 2 + 0 \times 4 \\ 0 \times 1 + 1 \times 3 & 0 \times 2 + 1 \times 4 \end{pmatrix} = \begin{pmatrix} 1 & 2 \\ 3 & 4 \end{pmatrix}

此时 R ( C ) = 2 R(C) = 2 ,而 m i n { R ( A ) , R ( B ) } = 2 min\{ R(A),R(B)\} = 2 R ( C ) = m i n { R ( A ) , R ( B ) } R(C) = min\{ R(A),R(B)\}

例2:降秩矩阵与满秩矩阵相乘

A = ( 1 1 2 2 ) A = \begin{pmatrix} 1 & 1 \\ 2 & 2 \end{pmatrix} R ( A ) = 1 R(A) = 1 ), B = ( 1 0 0 1 ) B = \begin{pmatrix} 1 & 0 \\ 0 & 1 \end{pmatrix} R ( B ) = 2 R(B) = 2 )。

计算乘积: C = A B = ( 1 × 1 + 1 × 0 1 × 0 + 1 × 1 2 × 1 + 2 × 0 2 × 0 + 2 × 1 ) = ( 1 1 2 2 ) C = AB = \begin{pmatrix} 1 \times 1 + 1 \times 0 & 1 \times 0 + 1 \times 1 \\ 2 \times 1 + 2 \times 0 & 2 \times 0 + 2 \times 1 \end{pmatrix} = \begin{pmatrix} 1 & 1 \\ 2 & 2 \end{pmatrix}

此时 R ( C ) = 1 R(C) = 1 ,而 m i n { R ( A ) , R ( B ) } = 1 min\{ R(A),R(B)\} = 1 R ( C ) = m i n { R ( A ) , R ( B ) } R(C) = min\{ R(A),R(B)\}

例3:两个降秩矩阵相乘

A = ( 1 2 2 4 ) A = \begin{pmatrix} 1 & 2 \\ 2 & 4 \end{pmatrix} R ( A ) = 1 R(A) = 1 ), B = ( 1 1 1 1 ) B = \begin{pmatrix} 1 & - 1 \\ - 1 & 1 \end{pmatrix} R ( B ) = 1 R(B) = 1 )。

计算乘积: C = A B = ( 1 × 1 + 2 × ( 1 ) 1 × ( 1 ) + 2 × 1 2 × 1 + 4 × ( 1 ) 2 × ( 1 ) + 4 × 1 ) = ( 1 1 2 2 ) C = AB = \begin{pmatrix} 1 \times 1 + 2 \times ( - 1) & 1 \times ( - 1) + 2 \times 1 \\ 2 \times 1 + 4 \times ( - 1) & 2 \times ( - 1) + 4 \times 1 \end{pmatrix} = \begin{pmatrix} - 1 & 1 \\ - 2 & 2 \end{pmatrix}

此时 R ( C ) = 1 R(C) = 1 ,而 m i n { R ( A ) , R ( B ) } = 1 min\{ R(A),R(B)\} = 1 R ( C ) = m i n { R ( A ) , R ( B ) } R(C) = min\{ R(A),R(B)\}

例4:乘积矩阵秩更小的情况

A = ( 1 0 0 0 ) A = \begin{pmatrix} 1 & 0 \\ 0 & 0 \end{pmatrix} R ( A ) = 1 R(A) = 1 ), B = ( 0 0 0 1 ) B = \begin{pmatrix} 0 & 0 \\ 0 & 1 \end{pmatrix} R ( B ) = 1 R(B) = 1 )。

计算乘积: C = A B = ( 1 × 0 + 0 × 0 1 × 0 + 0 × 1 0 × 0 + 0 × 0 0 × 0 + 0 × 1 ) = ( 0 0 0 0 ) C = AB = \begin{pmatrix} 1 \times 0 + 0 \times 0 & 1 \times 0 + 0 \times 1 \\ 0 \times 0 + 0 \times 0 & 0 \times 0 + 0 \times 1 \end{pmatrix} = \begin{pmatrix} 0 & 0 \\ 0 & 0 \end{pmatrix}

此时 R ( C ) = 0 R(C) = 0 ,而 m i n { R ( A ) , R ( B ) } = 1 min\{ R(A),R(B)\} = 1 R ( C ) < m i n { R ( A ) , R ( B ) } R(C) < min\{ R(A),R(B)\}

结论上述例子中,矩阵乘积 C = A B C = AB 的秩均不超过 A A B B 的秩的最小值,

验证了结论的正确性。