主要内容

高等代数教学辅导

4.2 线性映射和运算

建设中!

子节 4.2.1 主要知识点

定义 4.2.1.

\(U\)\(V\)是数域\(\mathbb{F} \)上的两个线性空间。若映射\(\phi: V\to U\)满足
  1. 对任意的\(\alpha_1\)\(\alpha_2\in V\),总有
    \begin{equation*} \phi(\alpha_1+\alpha_2) = \phi(\alpha_1)+\phi(\alpha_2) \end{equation*}
  2. 对任意的\(c\in \mathbb{F}\)\(\alpha\in V\),总有
    \begin{equation*} \phi(c\alpha) = c\phi(\alpha) \end{equation*}
则称\(\phi\)\(V\)\(U\)线性映射 。 特别的, 当\(V=U\)时,称\(\phi\)线性变换 ; 当\(U =\mathbb{F}\) 时, 称\(\phi\)线性函数

4.2.2.

\(V\)是数域\(\mathbb{F}\)上的\(n\)维线性空间,\(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n\)\(V\)的一组基,定义
\begin{equation*} \phi: V\to \mathbb{F}^n, \alpha\mapsto (a_1,a_2,\ldots,a_n)^T, \end{equation*}
其中\((a_1,a_2,\ldots,a_n)^T\)\(\alpha\)在基\(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n\)下的坐标,则\(\phi\)\(V\)\(\mathbb{F}^n\)的线性映射。

4.2.3.

线性空间\(V\)\(U\)的零映射\(0:\alpha\mapsto 0\)是线性映射。

4.2.4.

线性空间\(V\)的恒等映射\({\rm id}_V\)\(V\)的线性变换。

4.2.5.

\(V_1\)\(V_2\)\(V\)的子空间,\(V=V_1\oplus V_2\)。对\(i =1\)\(2\) 定义
\begin{gather*} \tau_i: V\to V_i,\ \alpha_1+\alpha_2\mapsto \alpha_i,\\ \sigma_1: V_1\to V,\ \alpha_1\mapsto \alpha_1+0,\\ \sigma_2: V_2\to V,\ \alpha_2\mapsto 0+\alpha_2, \end{gather*}
\(\tau_i\)\(\sigma_i\)是线性映射,称\(\tau_i\)投影映射\(\sigma_{i}\)嵌入映射

4.2.6.

\(A\in \mathbb{F}^{m\times n}\)。 验证如下定义的映射
\begin{equation*} \phi: \mathbb{F}^n\to \mathbb{F}^m,\ X\mapsto AX \end{equation*}
是线性映射。
  • \(r(A) = n \Leftrightarrow \phi\)为单线性映射;
  • \(r(A) = m \Leftrightarrow \phi\)为满线性映射。

4.2.7.

\(V=\mathbb{R}^2\)(实数域上二维向量空间),用\(T_\theta\)表示把\(V\)中每一个向量绕坐标原点旋转\(\theta\)角的变换,则\(T_\theta\)是一个线性变换。即
\begin{equation*} T_\theta: \mathbb{R}^2\to \mathbb{R}^2,\ \begin{pmatrix} x\\y \end{pmatrix} \mapsto \begin{pmatrix} x'\\y' \end{pmatrix} \end{equation*}
这里
\begin{equation*} \begin{pmatrix} x'\\y' \end{pmatrix} = \begin{pmatrix} \cos\theta & -\sin\theta\\ \sin\theta & \cos\theta \end{pmatrix} \begin{pmatrix} x\\y \end{pmatrix}. \end{equation*}
  • 线性映射 \(\phi\) 是单射 \(\Leftrightarrow\)\(\phi(\alpha)=0\),则 \(\alpha=0\)
  • 线性映射保持线性组合及关系式不变,即
    \begin{equation*} \phi(\sum_{i=1}^s k_i \alpha_i) =\sum_{i=1}^sk_i\phi(\alpha_i). \end{equation*}
  • 可作为 \(\phi\) 是线性映射的等价定义。
  1. 线性映射保零元、保负元、保线性组合。
  2. 线性映射保线性相关,但不保线性无关。单的线性映射保线性无关。

定义 4.2.11.

\(\phi\)\(\psi\)\(\mathbb{F}\)上线性空间\(V\to U\)的线性映射,则\(\phi\)\(\psi\) \(\phi+\psi\)定义为
\begin{equation*} \phi+\psi: V\to U,\ \alpha\mapsto\phi(\alpha)+\psi(\alpha) \end{equation*}
\(\phi+\psi\)也是\(V\to U\)的线性映射。
  • \(-\phi\)也称为\(\phi\)负映射

定义 4.2.13.

\(\phi\)\(\mathbb{F}\)上线性空间\(V\)\(U\)的线性映射, \(c\)\(\mathbb{F}\)的任意数,则\(c\)\(\phi\)数乘\(c\phi\)定义为
\begin{equation*} c\phi: V\to U,\ \alpha\mapsto c\phi(\alpha) \end{equation*}
\(c\phi\)也是\(V\to U\) 的线性映射。
\(V\)\(U\)\(\mathbb{F}\)上线性空间,\(\phi\in \mathcal{L}(V,U)\),分别取\(V\)\(U\)的基为:
\begin{equation*} V: \varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n;\quad U: \eta_1,\eta_2,\ldots,\eta_m. \end{equation*}
\(\begin{cases} \phi(\varepsilon_1)= {\color{red}a_{11}}\eta_1+{\color{red}a_{21}}\eta_2+{\color{red}\cdots}+{\color{red}a_{m1}}\eta_m\\ \phi(\varepsilon_2)= {\color{blue}a_{12}}\eta_1+{\color{blue}a_{22}}\eta_2+{\color{blue}\cdots}+{\color{blue}a_{m2}}\eta_m\\ {\hspace{3cm} \color{orange}\vdots}\\ \phi(\varepsilon_n)= {\color{green}a_{1n}}\eta_1+{\color{green}a_{2n}}\eta_2+{\color{green}\cdots}+{\color{green}a_{mn}}\eta_m \end{cases}\), 记
\begin{equation*} A =\begin{pmatrix} {\color{red}a_{11}} & {\color{blue}a_{12}} & {\color{orange}\cdots} & {\color{green}a_{1n}}\\ {\color{red}a_{21}} & {\color{blue}a_{22}} & {\color{orange}\cdots} & {\color{green}a_{2n}}\\ {\color{red}\vdots} & {\color{blue}{\vdots}} & {\color{orange}\ddots} & {\color{green}\vdots}\\ {\color{red}a_{m1}} & {\color{blue}a_{m2}} & {\color{orange}\cdots} & {\color{green}a_{mn}} \end{pmatrix}_{m\times n} \end{equation*}
则称矩阵\(A\)\(\phi\)\(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n\)\(\eta_1,\eta_2,\ldots,\eta_m\)下的矩阵或表示矩阵 。形式上记为
\begin{equation*} {\color{red}\phi(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n)=(\eta_1,\eta_2,\ldots,\eta_m)A}. \end{equation*}

定义 4.2.17.

\(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n\)为数域\(\mathbb{F}\)上线性空间\(V\)的一组基,\(\phi\)\(V\)上的线性变换。设
\begin{equation*} \phi(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n)=(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n)A_{n\times n}. \end{equation*}
矩阵\(A\)称为线性变换\(\phi\)在基\(\varepsilon_1,\varepsilon_2,\ldots,\varepsilon_n\)下的矩阵。

练习 4.2.2 练习

1.

判断下列映射是否是线性映射,并说明理由。
  1. \(\varphi:\mathbb{F}^3\rightarrow\mathbb{F}^2,\ (x,y,z)^T\mapsto (x-2z,y-z)^T\text{;}\)
  2. \(\varphi:\mathbb{F}^3\rightarrow\mathbb{F}^3,\ (x,y,z)^T\mapsto (x^2,y^2,z^2)^T\text{;}\)
  3. \(\varphi:\mathbb{F}^2\rightarrow\mathbb{F}^2,\ (x,y)^T\mapsto (x+a,y+b)^T\),其中\((a,b)^T\)\(\mathbb{F}^2\)中固定的向量;
  4. \(\varphi:\mathbb{F}^{m\times n}\rightarrow\mathbb{F}^{n\times m},\ X\mapsto X^T\text{;}\)
  5. \(\varphi:\mathbb{F}^{s\times t}\rightarrow\mathbb{F}^{m\times n},\ X\mapsto AXB\),其中\(A\in\mathbb{F}^{m\times s},B\in\mathbb{F}^{t\times n}\)是固定的矩阵;
  6. \(\varphi:\mathbb{F}^{n\times n}\rightarrow\mathbb{F}^{n\times n},\ X\mapsto XAX\),其中\(A\in\mathbb{F}^{n\times n}\)是固定的一个矩阵;
  7. \(\varphi :\mathbb{R}\rightarrow\mathbb{R}^+,\ x\mapsto 5^x\),其中\(\mathbb{R}^+\)作为\(\mathbb{R}\)上的线性空间,加法、数乘运算分别定义为\(a\oplus b=ab,\ k \odot a = a^k,\ \forall a,b\in\mathbb{R}^+,k\in\mathbb{R}\)。。

2.

\(A\in\mathbb{F}^{m\times n},B\in\mathbb{F}^{m\times s}\)是两个固定的矩阵,证明:映射
\begin{equation*} \varphi:\mathbb{F}^{n\times s}\rightarrow\mathbb{F}^{m\times s},\ X\mapsto AX+B \end{equation*}
是线性映射的充要条件为\(B=0\)

3.

\(\xi_1,\xi_2,\ldots,\xi_n\)是线性空间\(V\)的一个基,\(\varphi\)\(V\)\(U\)的线性映射。证明:\(\varphi\)是可逆映射的充要条件是\(\varphi(\xi_1),\varphi(\xi_2),\ldots,\varphi(\xi_n)\)\(U\)的一个基。

4.

\(V\)\(U\)是数域\(\mathbb{F}\)上的线性空间。\(\varphi\)\(V\to U\)的线性映射,\(V_1\)\(V_2\)\(V\)的子空间,\(U_1\)\(U\) 的子空间,\(S\)\(V\)的子集,证明:
  1. \(\varphi(V_1)\)\(U\)的子空间,\(\varphi^{-1}(U_1)\)\(V\)的子空间;
  2. \(\varphi(V_1+V_2)\)等于 \(\varphi(V_1)+\varphi(V_2)\)
  3. \(\varphi(\langle S\rangle)\)等于 \(\langle\varphi(S)\rangle\)
  4. 举例说明\(\varphi(V_1\bigcap V_2)\)未必等于 \(\varphi(V_1)\bigcap\varphi(V_2)\)
  5. 举例说明\(V_1\oplus V_2\not\Rightarrow\varphi(V_1)\oplus\varphi(V_2)\)

5.

\(V\)是数域\(\mathbb{F}\)上线性空间,\(\varphi :V\rightarrow V\)是线性变换,\(V_1\)\(V_2\)\(V\)的子空间,且\(V=V_1\oplus V_2\)。证明:\(\varphi\)是可逆映射的充分必要条件是
\begin{equation*} V=\varphi(V_1)\oplus \varphi(V_2). \end{equation*}

6.

\(\varphi:\mathbb{F}^4\rightarrow\mathbb{F}^3, \begin{pmatrix} a_1\\a_2\\a_3\\a_4 \end{pmatrix}\mapsto \begin{pmatrix} -a_1+a_2+2a_3+a_4\\-2a_2+a_3\\-a_1-a_2+3a_3+a_4 \end{pmatrix}\)
  1. 证明:\(\varphi\)是线性映射;
  2. \(\varphi\)在标准基\(\varepsilon_1,\varepsilon_2,\varepsilon_3,\varepsilon_4\)\(\varepsilon_1,\varepsilon_2,\varepsilon_3\)下的矩阵;
  3. \(\mathbb{F}^4\)内取一个基
    \begin{equation*} \alpha_1=(1,0,1,1)^T,\alpha_2=(0,1,0,1)^T,\alpha_3=(0,0,1,0)^T,\alpha_4=(0,0,2,1)^T, \end{equation*}
    又在\(\mathbb{F}^3\)内取一个基
    \begin{equation*} \beta_1=(1,1,1)^T,\beta_2=(1,0,-1)^T,\beta_3=(0,1,0)^T, \end{equation*}
    \(\varphi\)在基\(\alpha_1,\alpha_2,\alpha_3,\alpha_4\)\(\beta_1,\beta_2,\beta_3\)下的矩阵。

7.

两个函数
\begin{equation*} \xi_1=e^{ax}\cos bx,\quad \xi_2=e^{ax}\sin bx \end{equation*}
的所有实线性组合构成实数域上一个二维线性空间\(V\),求求导变换
\begin{equation*} \varphi:V\rightarrow V,f(x)\mapsto f'(x) \end{equation*}
在基\(\xi_1,\xi_2\)下的矩阵。

8.

定义
\begin{equation*} \varphi_1:\mathbb{F}^{2\times 2}\to\mathbb{F}^{2\times 2},\ X\mapsto AX, \end{equation*}
\begin{equation*} \varphi_2:\mathbb{F}^{2\times 2}\to\mathbb{F}^{2\times 2},\ X\mapsto XB, \end{equation*}
\begin{equation*} \varphi_3: \mathbb{F}^{2\times 2} \to \mathbb{F}_{2\times 2},\ X\mapsto AXB, \end{equation*}
其中\(A=\begin{pmatrix} 0 & 1\\ -1 & 0 \end{pmatrix}, B=\begin{pmatrix} 1&2\\3&4 \end{pmatrix}\text{,}\)\(\varphi_1,\varphi_2,\varphi_3\)在基\(E_{11},E_{12},E_{21},E_{22}\)下的矩阵。

9.

\(V\)是数域\(\mathbb{F}\)上的线性空间,\(\dim V=1\)\(\varphi\)\(V\)上的线性变换。证明:存在\(c\in\mathbb{F}\),使得对任意\(\alpha\in V\),都有\(\varphi(\alpha)=c\alpha\)

10.

\(W\)是有限维线性空间\(V\)的子空间,\(\varphi\)\(W\)\(V\)的线性映射,证明:\(\varphi\)可扩充为\(V\)上的线性变换\(\psi\),即存在线性变换\(\psi:V\rightarrow V\)使得
\begin{equation*} \psi (\alpha)=\varphi(\alpha),\ \forall\alpha\in W. \end{equation*}

11.

设线性映射\(\varphi:U\rightarrow V\)在基\(\alpha_1,\alpha_2\)\(\beta_1,\beta_2,\beta_3\)下的矩阵为
\begin{equation*} A=\begin{pmatrix} 2&0\\0&-2\\1&-1 \end{pmatrix}, \end{equation*}
\(U\)中向量\(\alpha\)在基\(\xi_1=\alpha_1,\xi_2=\alpha_1+\alpha_2\)下的坐标为\((2,3)^T\),求\(\varphi(\alpha)\)在基\(\eta_1=\beta_1+\beta_2,\eta_2=-2\beta_1-\beta_2,\eta_3=2\beta_2-\beta_3\)下的坐标。

12.

\(V,U,W\)是数域\(\mathbb{F}\)上的线性空间,\(\varphi\in\mathcal{L}(V,U),\psi\in\mathcal{L}(U,W)\)。证明:
  1. \(\psi\varphi\in\mathcal{L} (V,W)\)
  2. \(\varphi\)在基\(\xi_1,\xi_2,\ldots,\xi_n\)\(\eta_1,\eta_2,\ldots,\eta_m\)下的矩阵为\(A\)\(\psi\)在基\(\eta_1,\eta_2,\ldots,\eta_m\)\(\zeta_1,\zeta_2,\ldots,\zeta_s\)下的矩阵为\(B\),则\(\psi\varphi\)在基\(\xi_1,\xi_2,\ldots,\xi_n\)\(\zeta_1,\zeta_2,\ldots,\zeta_s\)下的矩阵为\(BA\)
  3. \(\varphi\)在基\(\xi_1,\xi_2,\ldots,\xi_n\)\(\eta_1,\eta_2,\ldots,\eta_m\)下的矩阵为\(A\),则\(\varphi\)是可逆映射的充要条件是\(A\)是可逆矩阵。此时,\(\varphi^{-1}\)在基\(\eta_1,\eta_2,\ldots,\eta_m\)\(\xi_1,\xi_2,\ldots,\xi_n\)下的矩阵为\(A^{-1}\)

13.

\(\varphi\)\(3\)维线性空间\(V\)上的线性变换,\(\xi_1,\xi_2,\xi_3\)\(V\)的一个基,
\begin{equation*} \varphi(\xi_1)=\xi_1,\varphi(\xi_2)=\xi_1+\xi_2,\varphi(\xi_3)=\xi_1+\xi_2+\xi_3, \end{equation*}
证明:\(\varphi\)是可逆映射,并求\(\varphi^{-1}\)