数学分析 隐函数定理及其应用(第18章)

一.隐函数
1.概念:

注:①这里只表示存在着定义在$I$上,值域包含于$J$的函数$f$，而不意味着$y$能用$x$的某一显式来表示

2.存在性


3.隐函数定理
(1)隐函数存在唯一性定理:

定理18.1:若函数$F(x,y)$满足:
①$F$在以$P_0(x_0,y_0)$为内点的某一区域$D\subset R^2$上连续
②$F(x_0,y_0)=0$(通常称为初始条件)
③$F$在$D$上存在连续的偏导数$F_y(x,y)$
④$F_y(x_0,y_0)\ne 0$
则
①存在点$P_0$的某邻域$U(P_0)\subset D$,在$U(P_0)$上方程$F(x,y)=0$唯一地决定了1个定义在某区间$(x_0-\alpha ,x_0+\alpha )$上的(隐)函数$y=f(x)$,使得当$x\in (x_0-\alpha ,x_0+\alpha )$时,$(x,f(x))\in U(P_0)$且$F(x,f(x))\equiv 0,f(x_0)=y_0$

(2)隐函数可微性定理:

定理18.2:设$F(x,y)$满足隐函数存在唯一性定理中的4个条件,又设在$D$上还存在连续的偏导数$F_x(x,y)$,则由方程(1)所确定的隐函数$y=f(x)$在其定义域$(x_0-\alpha ,x_0+\alpha )$上有连续偏导数,且$$f^{\prime }(x)=-\frac{F_x(x,y)}{F_y(x,y)}(5)$$

(3)隐函数极值问题:

(4)多元隐函数的唯一存在与连续可微定理:

定理18.3:若
①函数$F(x_1,x_2...x_n,y)$在以点$P_0(x_1^0,x_2^0...x_n^0,y^0)$为内点的区域$D\subset R^{n+1}$上连续
②$F(x_1^0,x_2^0...x_n^0,y^0)=0$
③偏导数$F_{x_1},F_{x_2}...F_{x_n},F_y$在$D$上存在且连续
④$F_y(x_1^0,x_2^0...x_n^0,y^0)\ne 0$
则
①存在点$P_0$的某邻域$U(P_0)\subset D$,在$U(P_0)$上方程$F(x_1,x_2...x_n,y)=0$唯一地确定了1个定义在$Q_0(x_1^0,x_2^0...x_n^0)$的某邻域$U(Q_0)\subset R^n$上的$n$元连续(隐)函数$y=f(x_1,x_2...x_n)$,使得当$(x_1,x_2...x_n)\in U(Q_0)$时,有$$(x_1,x_2...x_n,f(x_1,x_2...x_n))\in U(P_0)$$且$$F(x_1,x_2...x_n,f(x_1,x_2...x_n))\equiv 0,y^0=f((x_1^0,x_2^0...x_n^0)$$
②$y=f(x_1,x_2...x_n)$在$U(Q_0)$上有连续偏导数$f_{x_1},f_{x_2}...f_{x_n}$,且$$f_{x_1}=-\frac{F_{x_1}}{F_y},f_{x_2}=-\frac{F_{x_2}}{F_y}...f_{x_n}=-\frac{F_{x_n}}{F_y}$$

4.隐函数的反函数:

二.隐函数组
1.概念:


2.隐函数组定理
(1)函数行列式:

(2)隐函数组定理:

定理18.4:若
①$F(x,y,u,v),G(x,y,u,v)$在以点$P_0$为内点的区域$V\subset R^4$上连续
②$F(x_0,y_0,u_0,v_0)=G(x_0,y_0,u_0,v_0)$(初始条件)
③在$V$上$F,G$具有1阶连续偏导数
④$J=\frac{\partial (F,G)}{\partial (u,v)}$在点$P_0$处不等于0
则
①存在点$P_0$的某(4维空间)邻域$U(P_0)\subset V$,在$U(P_0)$上方程组(1)唯一地确定了定义在点$Q_0(x_0,y_0)$的某(2维空间)邻域$U(Q_0)$上的2个二元隐函数$$u=f(x,y),v=g(x,y)$$使得$u_0=f(x_0,y_0),v_0=g(x_0,y_0)$,且当$(x,y)\in U(Q_0)$时,有$$\begin{gathered} (x,y,f(x,y),g(x,y))\in U(P_0) \\ F(x,y,f(x,y),g(x,y))\equiv 0 \\ G(x,y,f(x,y),g(x,y))\equiv 0 \end{gathered}$$②$f(x,y),g(x,y)$在$U(Q_0)$上连续
③$f(x,y),g(x,y)$在$U(Q_0)$上有1阶连续偏导数,且$$\begin{array}{c}\frac{\partial u}{\partial x}=-\frac{1}{J}\frac{\partial (F,G)}{\partial (x,v)},\frac{\partial v}{\partial x}=-\frac{1}{J}\frac{\partial (F,G)}{\partial (u,x)}\\ \frac{\partial u}{\partial y}=-\frac{1}{J}\frac{\partial (F,G)}{\partial (y,v)},\frac{\partial v}{\partial y}=-\frac{1}{J}\frac{\partial (F,G)}{\partial (u,y)}\end{array}(5)$$

3.反函数组与坐标变换
(1)反函数组:


(2)反函数组定理:

定理18.5:设函数组(9)及其1阶偏导数在某区域$D\subset R^2$上连续,点$P_0(x_0,y_0)$是$D$的内点,且$$u_0=u(x_0,y_0),v_0=v(x_0,y_0),\frac{\partial (u,v)}{\partial (x,y)}{|}_{P_0}\ne 0$$则在点$P_0^{\prime }(u_0,v_0)$的某邻域$U(P_0^{\prime })$上存在唯一的1组反函数(10),使得$x_0=x(u_0,v_0),y_0=y(u_0,v_0)$,且当$(u,v)\in U(P_0^{\prime })$时,有$$(x(u,v),y(u,v))\in U(P_0)$$以及恒等式(11);此外,反函数组(10)在$U(P_0^{\prime })$上存在连续的1阶偏导数,且$$\begin{array}{c}\frac{\partial x}{\partial u}=\frac{\partial v}{\partial y}/\frac{\partial (u,v)}{\partial (x,y)},\frac{\partial x}{\partial v}=-\frac{\partial u}{\partial y}/\frac{\partial (u,v)}{\partial (x,y)}\\ \frac{\partial y}{\partial u}=-\frac{\partial v}{\partial x}/\frac{\partial (u,v)}{\partial (x,y)},\frac{\partial y}{\partial v}=\frac{\partial u}{\partial x}/\frac{\partial (u,v)}{\partial (x,y)}\end{array}(13)$$

(3)坐标变换:





三.几何应用
1.平面曲线的切线与法线:


2.空间曲线的切线与法平面




3.曲面的切平面与法线:

四.条件极值
1.条件极值问题:

2.拉格朗日乘数法:

定理18.6:设在条件(2)的限制下,求函数(3)的极值问题,其中$f,{\phi }_k(k=1,2...m)$在区域$D$上有连续的1阶偏导数;若$D$的内点$P_0(x_1^{(0)}...x_n^{(0)})$是上述问题的极值点,且雅可比矩阵$$\left[\begin{array}{ccc}\frac{\partial {\phi }_1}{\partial x_1}& ...& \frac{\partial {\phi }_1}{\partial x_n}\\ ...& & ...\\ \frac{\partial {\phi }_m}{\partial x_1}& ...& \frac{\partial {\phi }_m}{\partial x_n}\end{array}\right](13)$$的秩为$m$,则$\exists m$个常数${\lambda }_1^{(0)}...{\lambda }_m^{(0)}$,使得$(x_1^{(0)}...x_n^{(0)},{\lambda }_1^{(0)}...{\lambda }_m^{(0)})$为拉格朗日函数(12)的稳定点,即$(x_1^{(0)}...x_n^{(0)},{\lambda }_1^{(0)}...{\lambda }_m^{(0)})$为$n+m$个方程$$\{\begin{array}{l}{L}_{x1}=\frac{\partial f}{\partial x_1}+\sum _{k=1}^m{\lambda }_k\frac{\partial {\phi }_k}{\partial x_1}=0\\ ...\\ L_{x_n}=\frac{\partial f}{\partial x_n}+\sum _{k=1}^m{\lambda }_k\frac{\partial {\phi }_k}{\partial x_n}=0\\ L_{{\lambda }_1}={\phi }_1(x_1...x_n)\\ ...\\ L_{{\lambda }_m}={\phi }_m(x_1...x_n)\end{array}$$的解