93 lines
5.2 KiB
TeX
93 lines
5.2 KiB
TeX
\chapter{曲面的表示与逼近}
|
||
\section{曲面的表示}
|
||
\subsection{曲面的显式表示}
|
||
设有界闭区域$D \subset \realnum^2$，函数$f:D \to \ndreal$连续，我们称函数$f$的图像
|
||
\[G(f) = \{(x, y, f(x, y)) \in \realnum^3 \mid (x, y) \in D\}\]
|
||
为一张曲面，它展布在$D$上，称这曲面是由显式方程
|
||
\[z = f(x, y), (x, y) \in D\]
|
||
所确定的。
|
||
|
||
\subsection{曲面的隐式表示}
|
||
设三元函数$F$定义在区域$D \subset \realnum^3$，区域$D$中所有满足方程
|
||
\[F(x, y, z) = 0\tag{$\ast$} \label{曲面隐式表示}\]
|
||
的点集组成一张曲面，称为由方程\eqref{曲面隐式表示}所确定的隐式曲面。
|
||
|
||
\subsection{曲面的参数表示}
|
||
设$\bvec{f}: D \to \realnum^3$，$D \subset \realnum^2$是平面区域。则集合
|
||
\[S = \{(x, y, z) \mid (x, y, z) = \bvec{f}(u, v), (u, v) \in D\} = f(D)\]
|
||
称为$\realnum^3$空间中的一个曲面，$\bvec{f}(u, v)$称为曲面的$S$的参数表示。
|
||
|
||
\section{曲面的法向与切平面}
|
||
\subsection{有显式表示的曲面的切平面与法向量}
|
||
设曲面$S$有显式表示$z = f(x, y)$，令$z_0 = f(x_0, y_0)$，则$P = (x_0, y_0, z_0) \in S$，且$S$在$P$点的切平面方程为
|
||
\[z = z_0 + \frac{\partial f}{\partial x} (x_0, y_0) (x - x_0) + \frac{\partial f}{\partial y} (x_0, y_0) (y - y_0)\]
|
||
该平面的法向量
|
||
\[\bvec{n} = \pm \left(\frac{\partial f}{\partial x}(x_0, y_0), \frac{\partial f}{\partial y}(x_0, y_0), -1\right)\]
|
||
因此切平面方程还可以写成向量内积的形式：
|
||
\[\brak{\bvec{n}, \bvec{r} - \bvec{r}_0} = 0\]
|
||
其中
|
||
\[\bvec{r} = (x, y, z), \bvec{r}_0 = (x_0, y_0, z_0)\eqper\]
|
||
|
||
\subsection{隐式曲面的切平面与法向量}
|
||
设曲面$S$有隐式表示$F(x, y, z) = 0$，取$P = (x_0, y_0, z_0) \in S$，即$F(x_0, y_0, z_0) = 0$，不妨令$F \in C^1$且$\dfrac{\partial F}{\partial z}(x_0, y_0, z_0) \neq 0$，则根据隐函数定理，$P$点附近$S$有显示表示$z = z(x, y)$，切平面方程为
|
||
\[z = z_0 + \frac{\partial z}{\partial x}(x_0, y_0) (x - x_0) + \frac{\partial z}{\partial y}(x_0, y_0) (y - y_0)\]
|
||
其中
|
||
\[\frac{\partial z}{\partial x} = -\frac{\dfrac{\partial F}{\partial x}}{\dfrac{\partial F}{\partial z}}, \frac{\partial z}{\partial y} = -\frac{\dfrac{\partial F}{\partial y}}{\dfrac{\partial F}{\partial z}}\]
|
||
带入切平面方程，整理得
|
||
\[\frac{\partial F}{\partial x}(P) (x - x_0) + \frac{\partial F}{\partial y}(P) (y - y_0) + \frac{\partial F}{\partial z}(P) (z - z_0) = 0\]
|
||
因此法向量
|
||
\[\bvec{n} = \pm \gra F(P)\]
|
||
写成向量内积的形式
|
||
\[\brak{\bvec{n}, \bvec{r} - \bvec{r}_0} = 0\eqper\]
|
||
|
||
\subsection{参数曲面的切平面与法向量}
|
||
设曲面$S$有参数表示$\bvec{r} = \bvec{r}(u, v)$，$(u, v) \in D$。写成分量形式：$x = x(u, v), y = y(u, v), z = z(u, v)$。
|
||
|
||
令$(u_0, v_0) \in D$对应$P = (x_0, y_0, z_0) \in S$。只考虑$u$变化时对应的曲线：$\bvec{r} = \bvec{r}(u, v_0)$，切向为$\dfrac{\partial \bvec{r}}{\partial u}(u, v_0)$；类似地只考虑$v$变化时对应的曲线切向为$\dfrac{\partial \bvec{r}}{\partial v}(u_0, v)$。这两个法向量都应在$S$在$(u_0, v_0)$的切平面内，因此$S$在$P$点切平面的法向量$\bvec{n}$满足$\bvec{n} \perp \dfrac{\partial \bvec{r}}{\partial u}(u_0, v_0)$与$\bvec{n} \perp \dfrac{\partial \bvec{r}}{\partial v}(u_0, v_0)$。
|
||
|
||
综合起来，得到
|
||
\[\bvec{n} \parallel \dfrac{\partial \bvec{r}}{\partial u}(u_0, v_0) \times \dfrac{\partial \bvec{r}}{\partial v}(u_0, v_0)\]
|
||
|
||
进一步假设$\dfrac{\partial \bvec{r}}{\partial u}(u_0, v_0) \times \dfrac{\partial \bvec{r}}{\partial v}(u_0, v_0) \neq \bvec{0}$即两向量不共线，则可取$S$在$P$点法向$\bvec{n} = \dfrac{\partial \bvec{r}}{\partial u}(u_0, v_0) \times \dfrac{\partial \bvec{r}}{\partial v}(u_0, v_0)$，得到切平面方程
|
||
\[\brak{\dfrac{\partial \bvec{r}}{\partial u}(u_0, v_0) \times \dfrac{\partial \bvec{r}}{\partial v}(u_0, v_0), \bvec{r} - \bvec{r}_0} = 0\]
|
||
利用行列式也可以得到切平面方程
|
||
\[\begin{vmatrix}
|
||
x - x_0 & y - y_0 & z - z_0\\
|
||
D_u x(u_0, v_0) & D_u y(u_0, v_0) & D_u z(u_0, v_0)\\
|
||
D_v x(u_0, v_0) & D_v y(y_0, v_0) & D_v z(u_0, v_0)
|
||
\end{vmatrix}
|
||
= 0\]
|
||
与法向量
|
||
\[\bvec{n} = \dfrac{\partial \bvec{r}}{\partial u}(u_0, v_0) \times \dfrac{\partial \bvec{r}}{\partial v}(u_0, v_0) =
|
||
\begin{vmatrix}
|
||
\bvec{e}_1 & \bvec{e}_2 & \bvec{e}_3\\
|
||
D_u x & D_u y & D_u z\\
|
||
D_v x & D_v y & D_v z
|
||
\end{vmatrix}\eqper\]
|
||
|
||
同时，记$D_u \bvec{r} \times D_v \bvec{r} = (A, B, C)$，由定义
|
||
\[A = \begin{vmatrix}
|
||
D_u y & D_u z\\
|
||
D_v y & D_v z
|
||
\end{vmatrix},
|
||
B = \begin{vmatrix}
|
||
D_u z & D_u x\\
|
||
D_v z & D_v x
|
||
\end{vmatrix},
|
||
C = \begin{vmatrix}
|
||
D_u x & D_u y\\
|
||
D_v x & D_v y
|
||
\end{vmatrix}\]
|
||
引入第一基本量记号
|
||
\[E = \brak{D_u \bvec{r}, D_u \bvec{r}}, F = \brak{D_u \bvec{r}, D_v \bvec{r}}, G = \brak{D_v \bvec{r}, D_v \bvec{r}}\eqper\]
|
||
|
||
\section{曲线的切向量}
|
||
对曲线$\bvec{r} = \bvec{r}(t)$，在$\bvec{r}_0 = \bvec{r}(t_0)$处的切向量为
|
||
\[\deriv{\bvec{r}}(t_0) = \left(\deriv{x}(t_0), \deriv{y}(t_0) \deriv{z}(t_0)\right)\eqper\]
|
||
|
||
对曲线
|
||
\(\left\{\begin{aligned}
|
||
& F(x, y, z) = 0\\
|
||
& G(x, y, z) = 0
|
||
\end{aligned}\right.\)
|
||
即曲面$F(x, y, z) = 0$与$G(x, y, z) = 0$的交线，在$\bvec{r}_0 = (x_0, y_0, z_0)$处的切线为$\gra F(\bvec{r}_0) \times \gra G(\bvec{r}_0)$。 |