(二)协(逆)变基矢

在一般的曲线坐标系中: d r ⃗ = ∂ r ⃗ ∂ x i d x i d\vec{r}=\frac{\partial\vec{r}}{\partial x^i}d x^i dr =xir dxi定义协变基矢/自然基矢 g ⃗ i = ∂ r ⃗ ∂ x i \vec{g}_i=\frac{\partial\vec{r}}{\partial x^i} g i=xir 显然,协变基矢 g ⃗ i \vec{g}_i g i与坐标线 x i x_i xi的相切,指向其正方向,大小为坐标 x i x_i xi变化单位增量时前后两点的距离。借助直角坐标系作为参考: r ⃗ ( x 1 , x 2 , x 3 ) = x ( x 1 , x 2 , x 3 ) i ⃗ + y ( x 1 , x 2 , x 3 ) j ⃗ + z ( x 1 , x 2 , x 3 ) k ⃗ \vec{r}(x^1,x^2,x^3)=x(x^1,x^2,x^3)\vec{i}+y(x^1,x^2,x^3)\vec{j}+z(x^1,x^2,x^3)\vec{k} r (x1,x2,x3)=x(x1,x2,x3)i +y(x1,x2,x3)j +z(x1,x2,x3)k 则: { g ⃗ 1 = ∂ x ∂ x 1 i ⃗ + ∂ y ∂ x 1 j ⃗ + ∂ z ∂ x 1 k ⃗ = [ ∂ x ∂ x 1 ∂ y ∂ x 1 ∂ z ∂ x 1 ] T g ⃗ 2 = ∂ x ∂ x 2 i ⃗ + ∂ y ∂ x 2 j ⃗ + ∂ z ∂ x 2 k ⃗ = [ ∂ x ∂ x 2 ∂ y ∂ x 2 ∂ z ∂ x 2 ] T g ⃗ 3 = ∂ x ∂ x 3 i ⃗ + ∂ y ∂ x 3 j ⃗ + ∂ z ∂ x 3 k ⃗ = [ ∂ x ∂ x 3 ∂ y ∂ x 3 ∂ z ∂ x 3 ] T \begin{cases} \vec{g}_1=\dfrac{\partial x}{\partial x^1}\vec{i}+\dfrac{\partial y}{\partial x^1}\vec{j}+\dfrac{\partial z}{\partial x^1}\vec{k}=\begin{bmatrix}\dfrac{\partial x}{\partial x^1}&\dfrac{\partial y}{\partial x^1}&\dfrac{\partial z}{\partial x^1}\end{bmatrix}^T\\ \\ \vec{g}_2=\dfrac{\partial x}{\partial x^2}\vec{i}+\dfrac{\partial y}{\partial x^2}\vec{j}+\dfrac{\partial z}{\partial x^2}\vec{k}=\begin{bmatrix}\dfrac{\partial x}{\partial x^2}&\dfrac{\partial y}{\partial x^2}&\dfrac{\partial z}{\partial x^2}\end{bmatrix}^T\\ \\ \vec{g}_3=\dfrac{\partial x}{\partial x^3}\vec{i}+\dfrac{\partial y}{\partial x^3}\vec{j}+\dfrac{\partial z}{\partial x^3}\vec{k}=\begin{bmatrix}\dfrac{\partial x}{\partial x^3}&\dfrac{\partial y}{\partial x^3}&\dfrac{\partial z}{\partial x^3}\end{bmatrix}^T \end{cases} g 1=x1xi +x1yj +x1zk =[x1xx1yx1z]Tg 2=x2xi +x2yj +x2zk =[x2xx2yx2z]Tg 3=x3xi +x3yj +x3zk =[x3xx3yx3z]T
由于 d e t ( [ g ⃗ 1 g ⃗ 2 g ⃗ 3 ] ) = g ⃗ 1 ∙ ( g ⃗ 2 × g ⃗ 3 ) = g = ∣ ∂ x ∂ x 1 ∂ x ∂ x 2 ∂ x ∂ x 3 ∂ y ∂ x 1 ∂ y ∂ x 2 ∂ y ∂ x 3 ∂ z ∂ x 1 ∂ y ∂ x 2 ∂ y ∂ x 3 ∣ = d e t ( ∂ ( x , y , z ) ∂ ( x 1 , x 2 , x 3 ) ) ≠ 0 (不为零由曲线坐标系的定义得到) det(\begin{bmatrix}\vec{g}_1&\vec{g}_2&\vec{g}_3\end{bmatrix}) =\vec{g}_1\bullet(\vec{g}_2\times\vec{g}_3)=\sqrt{g} =\begin{vmatrix} \dfrac{\partial x}{\partial x^1}&\dfrac{\partial x}{\partial x^2}&\dfrac{\partial x}{\partial x^3}\\ \\ \dfrac{\partial y}{\partial x^1}&\dfrac{\partial y}{\partial x^2}&\dfrac{\partial y}{\partial x^3}\\ \\ \dfrac{\partial z}{\partial x^1}&\dfrac{\partial y}{\partial x^2}&\dfrac{\partial y}{\partial x^3} \end{vmatrix} =det\left(\dfrac{\partial(x,y,z)}{\partial(x^1,x^2,x^3)}\right)\ne0(不为零由曲线坐标系的定义得到) det([g 1g 2g 3])=g 1(g 2×g 3)=g = x1xx1yx1zx2xx2yx2yx3xx3yx3y =det((x1,x2,x3)(x,y,z))=0(不为零由曲线坐标系的定义得到)
则说明向量组 { g ⃗ 1 , g ⃗ 2 , g ⃗ 2 } \{\vec{g}_1,\vec{g}_2,\vec{g}_2\} {g 1,g 2,g 2}线性无关,其中 g \sqrt{g} g 称作体积单元 ,那么该向量组在空间中任意一点可构成一个坐标架 (切标架)。考虑某个矢量 v ⃗ \vec{v} v 在协变基矢构成的局部基上的分解: v ⃗ = p i g ⃗ i = p 1 g ⃗ 1 + p 2 g ⃗ 2 + p 3 g ⃗ 3 \vec{v}=p^i\vec{g}_i=p^1\vec{g}_1+p^2\vec{g}_2+p^3\vec{g}_3 v =pig i=p1g 1+p2g 2+p3g 3为求解矢量的逆变分量 p i p^i pi ,有: { p 1 ( g ⃗ 1 , g ⃗ 1 ) + p 2 ( g ⃗ 2 , g ⃗ 1 ) + p 3 ( g ⃗ 3 , g ⃗ 1 ) = ( v ⃗ , g ⃗ 1 ) p 1 ( g ⃗ 1 , g ⃗ 2 ) + p 2 ( g ⃗ 2 , g ⃗ 2 ) + p 3 ( g ⃗ 3 , g ⃗ 2 ) = ( v ⃗ , g ⃗ 2 ) p 1 ( g ⃗ 1 , g ⃗ 3 ) + p 2 ( g ⃗ 2 , g ⃗ 3 ) + p 3 ( g ⃗ 3 , g ⃗ 3 ) = ( v ⃗ , g ⃗ 1 ) \begin{cases} p^1(\vec{g}_1,\vec{g}_1)+p^2(\vec{g}_2,\vec{g}_1)+p^3(\vec{g}_3,\vec{g}_1)=(\vec{v},\vec{g}_1)\\ p^1(\vec{g}_1,\vec{g}_2)+p^2(\vec{g}_2,\vec{g}_2)+p^3(\vec{g}_3,\vec{g}_2)=(\vec{v},\vec{g}_2)\\ p^1(\vec{g}_1,\vec{g}_3)+p^2(\vec{g}_2,\vec{g}_3)+p^3(\vec{g}_3,\vec{g}_3)=(\vec{v},\vec{g}_1) \end{cases} p1(g 1,g 1)+p2(g 2,g 1)+p3(g 3,g 1)=(v ,g 1)p1(g 1,g 2)+p2(g 2,g 2)+p3(g 3,g 2)=(v ,g 2)p1(g 1,g 3)+p2(g 2,g 3)+p3(g 3,g 3)=(v ,g 1)上述非齐次线性方程组总是有解的,因为其系数矩阵可逆,即:
∣ ( g ⃗ 1 , g ⃗ 1 ) ( g ⃗ 2 , g ⃗ 1 ) ( g ⃗ 3 , g ⃗ 1 ) ( g ⃗ 1 , g ⃗ 2 ) ( g ⃗ 2 , g ⃗ 2 ) ( g ⃗ 3 , g ⃗ 2 ) ( g ⃗ 1 , g ⃗ 3 ) ( g ⃗ 2 , g ⃗ 3 ) ( g ⃗ 3 , g ⃗ 3 ) ∣ = [ g ⃗ 1 ∙ ( g ⃗ 2 × g ⃗ 3 ) ] 2 ≠ 0 \begin{vmatrix} (\vec{g}_1,\vec{g}_1)&(\vec{g}_2,\vec{g}_1)&(\vec{g}_3,\vec{g}_1)\\ \\ (\vec{g}_1,\vec{g}_2)&(\vec{g}_2,\vec{g}_2)&(\vec{g}_3,\vec{g}_2)\\ \\ (\vec{g}_1,\vec{g}_3)&(\vec{g}_2,\vec{g}_3)&(\vec{g}_3,\vec{g}_3) \end{vmatrix} = [\vec{g}_1\bullet(\vec{g}_2\times\vec{g}_3)]^2\ne0 (g 1,g 1)(g 1,g 2)(g 1,g 3)(g 2,g 1)(g 2,g 2)(g 2,g 3)(g 3,g 1)(g 3,g 2)(g 3,g 3) =[g 1(g 2×g 3)]2=0但上述求解过程较麻烦,为更方便地求解逆变分量引入逆变基矢 g ⃗ j \vec{g}^j g j,要求其与协变基矢满足对偶关系 ( g ⃗ i , g ⃗ j ) = δ i j = { 1 ( i = j ) 0 ( i ≠ j ) (\vec{g}_i,\vec{g}^j)=\delta_i^j=\begin{cases}1\quad(i=j)\\0\quad(i\ne j)\end{cases} (g i,g j)=δij={1(i=j)0(i=j)那么 p i = ( v ⃗ , g ⃗ i ) p^i=(\vec{v},\vec{g}^i) pi=(v ,g i)接下来根据对偶关系由协变基矢求解逆变基矢:
方法一:
参考直角坐标系讨论,设 g ⃗ j = g 1 j i ⃗ + g 2 j j ⃗ + g 3 j k ⃗ = [ g 1 j g 2 j g 3 j ] T ( j = 1 , 2 , 3 ) \vec{g}^j=g^j_1\vec{i}+g^j_2\vec{j}+g^j_3\vec{k}=\begin{bmatrix}g^j_1&g^j_2&g^j_3\end{bmatrix}^T\quad(j=1,2,3) g j=g1ji +g2jj +g3jk =[g1jg2jg3j]T(j=1,2,3)不妨以 j = 1 j=1 j=1 时进行讨论:
{ ( g ⃗ 1 , g ⃗ 1 ) = g 1 1 ∂ x ∂ x 1 + g 2 1 ∂ y ∂ x 1 + g 3 1 ∂ z ∂ x 1 = 1 ( g ⃗ 1 , g ⃗ 2 ) = g 1 1 ∂ x ∂ x 2 + g 2 1 ∂ y ∂ x 2 + g 3 1 ∂ z ∂ x 2 = 0 ( g ⃗ 1 , g ⃗ 1 ) = g 1 1 ∂ x ∂ x 3 + g 2 1 ∂ y ∂ x 3 + g 3 1 ∂ z ∂ x 3 = 0 \begin{cases} (\vec{g}^1,\vec{g}_1)=g^1_1\dfrac{\partial x}{\partial x^1}+g^1_2\dfrac{\partial y}{\partial x^1}+g^1_3\dfrac{\partial z}{\partial x^1}=1\\ \\ (\vec{g}^1,\vec{g}_2)=g^1_1\dfrac{\partial x}{\partial x^2}+g^1_2\dfrac{\partial y}{\partial x^2}+g^1_3\dfrac{\partial z}{\partial x^2}=0\\ \\ (\vec{g}^1,\vec{g}_1)=g^1_1\dfrac{\partial x}{\partial x^3}+g^1_2\dfrac{\partial y}{\partial x^3}+g^1_3\dfrac{\partial z}{\partial x^3}=0 \end{cases} (g 1,g 1)=g11x1x+g21x1y+g31x1z=1(g 1,g 2)=g11x2x+g21x2y+g31x2z=0(g 1,g 1)=g11x3x+g21x3y+g31x3z=0由于系数矩阵满足: d e t ( ∂ ( x , y , z ) ∂ ( x 1 , x 2 , x 3 ) T ) = d e t ( ∂ ( x , y , z ) ∂ ( x 1 , x 2 , x 3 ) ) ≠ 0 , ∂ ( x , y , z ) ∂ ( x 1 , x 2 , x 3 ) ∂ ( x 1 , x 2 , x 3 ) ∂ ( x , y , z ) = E det\left(\dfrac{\partial (x,y,z)}{\partial (x^1,x^2,x^3)}^T\right)=det\left(\dfrac{\partial (x,y,z)}{\partial (x^1,x^2,x^3)}\right)\ne0,\dfrac{\partial (x,y,z)}{\partial (x^1,x^2,x^3)}\dfrac{\partial (x^1,x^2,x^3)}{\partial (x,y,z)}=E det((x1,x2,x3)(x,y,z)T)=det((x1,x2,x3)(x,y,z))=0(x1,x2,x3)(x,y,z)(x,y,z)(x1,x2,x3)=E
[ g ⃗ 1 1 g ⃗ 2 1 g ⃗ 3 1 ] = [ ∂ x 1 ∂ x ∂ x 2 ∂ x ∂ x 3 ∂ x ∂ x 1 ∂ y ∂ x 2 ∂ y ∂ x 3 ∂ y ∂ x 1 ∂ z ∂ x 2 ∂ z ∂ x 3 ∂ z ] [ 1 0 0 ] \begin{bmatrix} \vec{g}^1_1\\ \\ \vec{g}^1_2\\ \\ \vec{g}^1_3 \end{bmatrix} = \begin{bmatrix} %%%%%%%%%% \dfrac{\partial x^1}{\partial x}&\dfrac{\partial x^2}{\partial x}&\dfrac{\partial x^3}{\partial x}\\ \\ \dfrac{\partial x^1}{\partial y}&\dfrac{\partial x^2}{\partial y}&\dfrac{\partial x^3}{\partial y}\\ \\ \dfrac{\partial x^1}{\partial z}&\dfrac{\partial x^2}{\partial z}&\dfrac{\partial x^3}{\partial z} \end{bmatrix} \begin{bmatrix} %%%%%%%%%%% 1\\ \\ 0\\ \\ 0 \end{bmatrix} g 11g 21g 31 = xx1yx1zx1xx2yx2zx2xx3yx3zx3 100 同理可求得 g ⃗ 2 , g ⃗ 3 \vec{g}^2,\vec{g}^3 g 2,g 3,即
[ g ⃗ 1 g ⃗ 2 g ⃗ 3 ] = [ g ⃗ 1 1 g ⃗ 1 2 g ⃗ 1 3 g ⃗ 2 1 g ⃗ 2 2 g ⃗ 2 3 g ⃗ 3 1 g ⃗ 3 2 g ⃗ 3 3 ] = [ ∂ x 1 ∂ x ∂ x 2 ∂ x ∂ x 3 ∂ x ∂ x 1 ∂ y ∂ x 2 ∂ y ∂ x 3 ∂ y ∂ x 1 ∂ z ∂ x 2 ∂ z ∂ x 3 ∂ z ] [ 1 0 0 0 1 0 0 0 1 ] = ( ∂ ( x 1 , x 2 , x 3 ) ∂ ( x , y , z ) ) T \begin{bmatrix}\vec{g}^1&\vec{g}^2&\vec{g}^3\end{bmatrix}= \begin{bmatrix}%%%%%%%%%%%%%%% \vec{g}^1_1&\vec{g}^2_1&\vec{g}^3_1\\ \\ \vec{g}^1_2&\vec{g}^2_2&\vec{g}^3_2\\ \\ \vec{g}^1_3&\vec{g}^2_3&\vec{g}^3_3 \end{bmatrix} =\begin{bmatrix} %%%%%%%%%% \dfrac{\partial x^1}{\partial x}&\dfrac{\partial x^2}{\partial x}&\dfrac{\partial x^3}{\partial x}\\ \\ \dfrac{\partial x^1}{\partial y}&\dfrac{\partial x^2}{\partial y}&\dfrac{\partial x^3}{\partial y}\\ \\ \dfrac{\partial x^1}{\partial z}&\dfrac{\partial x^2}{\partial z}&\dfrac{\partial x^3}{\partial z} \end{bmatrix} \begin{bmatrix} %%%%%%%%%%% 1&0&0\\ \\ 0&1&0\\ \\ 0&0&1 \end{bmatrix} =\left(\dfrac{\partial (x^1,x^2,x^3)}{\partial (x,y,z)}\right)^T%%%%% [g 1g 2g 3]= g 11g 21g 31g 12g 22g 32g 13g 23g 33 = xx1yx1zx1xx2yx2zx2xx3yx3zx3 100010001 =((x,y,z)(x1,x2,x3))T换而言之, g ⃗ i = ▽ x i ( x , y , z ) ( i = 1 , 2 , 3 ) \vec{g}^i=\bigtriangledown x^i(x,y,z)\qquad(i=1,2,3) g i=xi(x,y,z)(i=1,2,3)这说明,逆变基矢为坐标面的梯度,方向沿某点坐标面的法向。
另外,由上述结论可知: d e t ( [ g ⃗ 1 g ⃗ 2 g ⃗ 3 ] ) ≠ 0 det(\begin{bmatrix}\vec{g}^1&\vec{g}^2&\vec{g}^3\end{bmatrix})\ne0 det([g 1g 2g 3])=0,说明逆变基矢构成的向量组也是线性无关的,可构成 R 3 \mathbb{R}^3 R3 中的一组基。

方法二:
由对偶关系: g ⃗ 1 \vec{g}^1 g 1 同时与 g ⃗ 2 , g ⃗ 3 \vec{g}_2,\vec{g}_3 g 2g 3 垂直,则应有: g ⃗ 1 = c ( g ⃗ 2 × g ⃗ 3 ) \vec{g}^1=c(\vec{g}_2\times\vec{g}_3) g 1=c(g 2×g 3)进一步利用对偶关系: 1 = ( g ⃗ 1 , g ⃗ 1 ) = c [ g ⃗ 1 ∙ ( g ⃗ 2 × g ⃗ 3 ) ] = c g ⟹ c = 1 g ⟹ g ⃗ 1 = g ⃗ 2 × g ⃗ 3 g 1=(\vec{g}^1,\vec{g}_1)=c[\vec{g}_1\bullet(\vec{g}_2\times\vec{g}_3)]=c\sqrt{g}\Longrightarrow c=\frac{1}{\sqrt{g}}\Longrightarrow\vec{g}^1=\dfrac{\vec{g}_2\times\vec{g}_3}{\sqrt{g}} 1=(g 1,g 1)=c[g 1(g 2×g 3)]=cg c=g 1g 1=g g 2×g 3同理 g ⃗ 2 = g ⃗ 3 × g ⃗ 1 g g ⃗ 3 = g ⃗ 1 × g ⃗ 2 g ( 注意:协变基矢叉乘的次序问题 ) \vec{g}^2=\dfrac{\vec{g}_3\times\vec{g}_1}{\sqrt{g}}\\ \quad\\ \vec{g}^3=\dfrac{\vec{g}_1\times\vec{g}_2}{\sqrt{g}}\\ \quad\\ (注意:协变基矢叉乘的次序问题) g 2=g g 3×g 1g 3=g g 1×g 2(注意:协变基矢叉乘的次序问题)方法三:
将逆变基矢在协变基矢构成的基上进行分解有:
[ g ⃗ 1 g ⃗ 2 g ⃗ 3 ] = [ g ⃗ 1 g ⃗ 2 g ⃗ 3 ] [ g 11 g 12 g 13 g 21 g 22 g 23 g 31 g 32 g 33 ] ⟺ g ⃗ j = g i j g ⃗ i ( i . j = 1 , 2 , 3 ) \begin{bmatrix}\vec{g}^1&\vec{g}^2&\vec{g}^3\end{bmatrix}= \begin{bmatrix}\vec{g}_1&\vec{g}_2&\vec{g}_3\end{bmatrix}%%%%%%%%% \begin{bmatrix}%%%%%%%%%%%%%%%% g^{11}&g^{12}&g^{13}\\ g^{21}&g^{22}&g^{23}\\ g^{31}&g^{32}&g^{33} \end{bmatrix}\Longleftrightarrow \vec{g}^j=g^{ij}\vec{g}_i\quad(i.j=1,2,3) [g 1g 2g 3]=[g 1g 2g 3] g11g21g31g12g22g32g13g23g33 g j=gijg i(i.j=1,2,3)由对偶关系: ( g ⃗ i , g ⃗ j ) = ( g ⃗ i , g k j g ⃗ k ) = g k j ( g ⃗ i , g ⃗ k ) = g k j δ k i = g i j ⟹ g i j = ( g ⃗ i , g ⃗ j ) = g j i (\vec{g}^i,\vec{g}^j)=(\vec{g}^i,g^{kj}\vec{g}_k)=g^{kj}(\vec{g}^i,\vec{g}_k)=g^{kj}\delta^i_k=g^{ij}\Longrightarrow g^{ij}=(\vec{g}^i,\vec{g}^j)=g^{ji} (g i,g j)=(g i,gkjg k)=gkj(g i,g k)=gkjδki=gijgij=(g i,g j)=gji同理,将协变基矢在逆变基矢构成的基上进行分解: [ g ⃗ 1 g ⃗ 2 g ⃗ 3 ] = [ g ⃗ 1 g ⃗ 2 g ⃗ 3 ] [ g 11 g 12 g 13 g 21 g 22 g 23 g 31 g 32 g 33 ] ⟺ g ⃗ j = g i j g ⃗ i ( i . j = 1 , 2 , 3 ) \begin{bmatrix}\vec{g}_1&\vec{g}_2&\vec{g}_3\end{bmatrix}= \begin{bmatrix}\vec{g}^1&\vec{g}^2&\vec{g}^3\end{bmatrix}%%%%%%%%% \begin{bmatrix}%%%%%%%%%%%%%%%% g_{11}&g_{12}&g_{13}\\ g_{21}&g_{22}&g_{23}\\ g_{31}&g_{32}&g_{33} \end{bmatrix}\Longleftrightarrow \vec{g}_j=g_{ij}\vec{g}^i\quad(i.j=1,2,3) [g 1g 2g 3]=[g 1g 2g 3] g11g21g31g12g22g32g13g23g33 g j=gijg i(i.j=1,2,3)由对偶关系: ( g ⃗ i , g ⃗ j ) = ( g ⃗ i , g k j g ⃗ k ) = g k j ( g ⃗ i , g ⃗ k ) = g k j δ i k = g i j ⟹ g i j = ( g ⃗ i , g ⃗ j ) = g j i (\vec{g}_i,\vec{g}_j)=(\vec{g}_i,g_{kj}\vec{g}^k)=g_{kj}(\vec{g}_i,\vec{g}^k)=g_{kj}\delta_i^k=g_{ij}\Longrightarrow g_{ij}=(\vec{g}_i,\vec{g}_j)=g_{ji} (g i,g j)=(g i,gkjg k)=gkj(g i,g k)=gkjδik=gijgij=(g i,g j)=gji显然, [ g 11 g 12 g 13 g 21 g 22 g 23 g 31 g 32 g 33 ] = [ g 11 g 12 g 13 g 21 g 22 g 23 g 31 g 32 g 33 ] − 1 ⟺ [ g i j ] = [ g i j ] − 1 \begin{bmatrix}%%%%%%%%%%%%%%%% g^{11}&g^{12}&g^{13}\\ g^{21}&g^{22}&g^{23}\\ g^{31}&g^{32}&g^{33} \end{bmatrix}= \begin{bmatrix}%%%%%%%%%%%%%%%% g_{11}&g_{12}&g_{13}\\ g_{21}&g_{22}&g_{23}\\ g_{31}&g_{32}&g_{33} \end{bmatrix}^{-1}\Longleftrightarrow [g^{ij}]=[g_{ij}]^{-1} g11g21g31g12g22g32g13g23g33 = g11g21g31g12g22g32g13g23g33 1[gij]=[gij]1由该方法求解逆变基矢时步骤是分析的逆过程,即:
(1) 由协变基矢求解度量张量的协变分量 g i j g_{ij} gij;
(2) 根据可逆关系由度量张量的协变分量求得度量张量的逆变分量 g i j g^{ij} gij;
(3) 由基矢量的指标升降关系 g ⃗ j = g i j g ⃗ i \vec{g}^j=g^{ij}\vec{g}_i g j=gijg i 求得逆变基矢。

对于协变基矢与逆变基矢构成的坐标架我们有如下结论: { g ⃗ 1 , g ⃗ 2 , g ⃗ 3 } \{\vec{g}_1,\vec{g}_2,\vec{g}_3\} {g 1,g 2,g 3}构成右(左)手坐标系,则 { g ⃗ 1 , g ⃗ 2 , g ⃗ 3 } \{\vec{g}^1,\vec{g}^2,\vec{g}^3\} {g 1,g 2,g 3}仍然构成右(左)手坐标系。 证明过程如下: d e t ( [ δ j i ] ) = ∣ δ 1 1 δ 2 1 δ 3 1 δ 1 2 δ 2 2 δ 3 2 δ 1 3 δ 2 3 δ 3 3 ∣ = ∣ ( g ⃗ 1 , g ⃗ 1 ) ( g ⃗ 1 , g ⃗ 2 ) ( g ⃗ 1 , g ⃗ 3 ) ( g ⃗ 2 , g ⃗ 1 ) ( g ⃗ 2 , g ⃗ 2 ) ( g ⃗ 2 , g ⃗ 3 ) ( g ⃗ 3 , g ⃗ 1 ) ( g ⃗ 3 , g ⃗ 2 ) ( g ⃗ 3 , g ⃗ 3 ) ∣ = [ g ⃗ 1 ∙ ( g ⃗ 2 × g ⃗ 3 ) ] [ g ⃗ 1 ∙ ( g ⃗ 2 × g ⃗ 3 ) ] = g   [ g ⃗ 1 ∙ ( g ⃗ 2 × g ⃗ 3 ) ] = 1 det([\delta^i_j]) =\begin{vmatrix}%%%%% \delta^1_1&\delta^1_2&\delta^1_3\\ \\ \delta^2_1&\delta^2_2&\delta^2_3\\ \\ \delta^3_1&\delta^3_2&\delta^3_3\\ \end{vmatrix} =\begin{vmatrix}%%%%% (\vec{g}^1,\vec{g}_1)&(\vec{g}^1,\vec{g}_2)&(\vec{g}^1,\vec{g}_3)\\ \\ (\vec{g}^2,\vec{g}_1)&(\vec{g}^2,\vec{g}_2)&(\vec{g}^2,\vec{g}_3)\\ \\ (\vec{g}^3,\vec{g}_1)&(\vec{g}^3,\vec{g}_2)&(\vec{g}^3,\vec{g}_3)\\ \end{vmatrix} =[\vec{g}^1\bullet(\vec{g}^2\times\vec{g}^3)][\vec{g}_1\bullet(\vec{g}_2\times\vec{g}_3)]%%%%% =\sqrt{g}\ [\vec{g}^1\bullet(\vec{g}^2\times\vec{g}^3)]=1 det([δji])= δ11δ12δ13δ21δ22δ23δ31δ32δ33 = (g 1,g 1)(g 2,g 1)(g 3,g 1)(g 1,g 2)(g 2,g 2)(g 3,g 2)(g 1,g 3)(g 2,g 3)(g 3,g 3) =[g 1(g 2×g 3)][g 1(g 2×g 3)]=g  [g 1(g 2×g 3)]=1则, g ⃗ 1 ∙ ( g ⃗ 2 × g ⃗ 3 ) = 1 g \vec{g}^1\bullet(\vec{g}^2\times\vec{g}^3)=\frac{1}{\sqrt{g}} g 1(g 2×g 3)=g 1上式给出了逆变基矢的混合积计算公式且说明它与协变基矢的混合积同号。(证毕)

借助上述逆变基矢混合积公式与对偶关系同样可获得与方法二类似的由逆变基矢求协变基矢的公式: g ⃗ 1 = g   ( g ⃗ 2 × g ⃗ 3 )   g ⃗ 2 = g   ( g ⃗ 3 × g ⃗ 1 )   g ⃗ 3 = g   ( g ⃗ 1 × g ⃗ 2 ) \vec{g}_1=\sqrt{g}\ (\vec{g}^2\times\vec{g}^3)\\ \ \\ \vec{g}_2=\sqrt{g}\ (\vec{g}^3\times\vec{g}^1)\\ \ \\ \vec{g}_3=\sqrt{g}\ (\vec{g}^1\times\vec{g}^2) g 1=g  (g 2×g 3) g 2=g  (g 3×g 1) g 3=g  (g 1×g 2)最后指出,根据度量协(逆)变分量与基矢量的关系式可得到如下关系式:
{ d e t ( [ g i j ] ) = 1 g d e t ( [ g i j ] ) = g \begin{cases}det([g^{ij}])=\frac{1}{g}\\ \\ det([g_{ij}])=g\end{cases} det([gij])=g1det([gij])=g
{ g 11 = ( ∂ x ∂ x 1 ) 2 + ( ∂ y ∂ x 1 ) 2 + ( ∂ z ∂ x 1 ) 2 g 22 = ( ∂ x ∂ x 2 ) 2 + ( ∂ y ∂ x 2 ) 2 + ( ∂ z ∂ x 2 ) 2 g 33 = ( ∂ x ∂ x 3 ) 2 + ( ∂ y ∂ x 3 ) 2 + ( ∂ z ∂ x 3 ) 2 g 12 = g 21 = ∂ x ∂ x 1 ∂ x ∂ x 2 + ∂ y ∂ x 1 ∂ y ∂ x 2 + ∂ z ∂ x 1 ∂ z ∂ x 2 g 13 = g 31 = ∂ x ∂ x 1 ∂ x ∂ x 3 + ∂ y ∂ x 1 ∂ y ∂ x 3 + ∂ z ∂ x 1 ∂ z ∂ x 3 g 23 = g 32 = ∂ x ∂ x 2 ∂ x ∂ x 3 + ∂ y ∂ x 2 ∂ y ∂ x 3 + ∂ z ∂ x 2 ∂ z ∂ x 3 { g 11 = ( ∂ x 1 ∂ x ) 2 + ( ∂ x 1 ∂ y ) 2 + ( ∂ x 1 ∂ z ) 2 g 22 = ( ∂ x 2 ∂ x ) 2 + ( ∂ x 2 ∂ y ) 2 + ( ∂ x 2 ∂ z ) 2 g 33 = ( ∂ x 3 ∂ x ) 2 + ( ∂ x 3 ∂ y ) 2 + ( ∂ x 3 ∂ z ) 2 g 12 = g 21 = ∂ x 1 ∂ x ∂ x 2 ∂ x + ∂ x 1 ∂ y ∂ x 2 ∂ y + ∂ x 1 ∂ z ∂ x 2 ∂ z g 13 = g 31 = ∂ x 1 ∂ x ∂ x 3 ∂ x + ∂ x 1 ∂ y ∂ x 3 ∂ y + ∂ x 1 ∂ z ∂ x 3 ∂ z g 23 = g 32 = ∂ x 2 ∂ x ∂ x 3 ∂ x + ∂ x 2 ∂ y ∂ x 3 ∂ y + ∂ x 2 ∂ z ∂ x 3 ∂ z \begin{cases} %%%%%%%%%%%%% g_{11}=\left(\dfrac{\partial x}{\partial x_1}\right)^2+\left(\dfrac{\partial y}{\partial x_1}\right)^2+\left(\dfrac{\partial z}{\partial x_1}\right)^2\\ \\ g_{22}=\left(\dfrac{\partial x}{\partial x_2}\right)^2+\left(\dfrac{\partial y}{\partial x_2}\right)^2+\left(\dfrac{\partial z}{\partial x_2}\right)^2\\ \\ g_{33}=\left(\dfrac{\partial x}{\partial x_3}\right)^2+\left(\dfrac{\partial y}{\partial x_3}\right)^2+\left(\dfrac{\partial z}{\partial x_3}\right)^2\\ \\ g_{12}=g_{21}=\dfrac{\partial x}{\partial x_1}\dfrac{\partial x}{\partial x_2}+\dfrac{\partial y}{\partial x_1}\dfrac{\partial y}{\partial x_2}+\dfrac{\partial z}{\partial x_1}\dfrac{\partial z}{\partial x_2}\\ \\ g_{13}=g_{31}=\dfrac{\partial x}{\partial x_1}\dfrac{\partial x}{\partial x_3}+\dfrac{\partial y}{\partial x_1}\dfrac{\partial y}{\partial x_3}+\dfrac{\partial z}{\partial x_1}\dfrac{\partial z}{\partial x_3}\\ \\ g_{23}=g_{32}=\dfrac{\partial x}{\partial x_2}\dfrac{\partial x}{\partial x_3}+\dfrac{\partial y}{\partial x_2}\dfrac{\partial y}{\partial x_3}+\dfrac{\partial z}{\partial x_2}\dfrac{\partial z}{\partial x_3} \end{cases} \qquad \begin{cases} %%%%%%%%%%%%%%%% g^{11}=\left(\dfrac{\partial x_1}{\partial x}\right)^2+\left(\dfrac{\partial x_1}{\partial y}\right)^2+\left(\dfrac{\partial x_1}{\partial z}\right)^2\\ \\ g^{22}=\left(\dfrac{\partial x_2}{\partial x}\right)^2+\left(\dfrac{\partial x_2}{\partial y}\right)^2+\left(\dfrac{\partial x_2}{\partial z}\right)^2\\ \\ g^{33}=\left(\dfrac{\partial x_3}{\partial x}\right)^2+\left(\dfrac{\partial x_3}{\partial y}\right)^2+\left(\dfrac{\partial x_3}{\partial z}\right)^2\\ \\ g^{12}=g^{21}=\dfrac{\partial x_1}{\partial x}\dfrac{\partial x_2}{\partial x}+\dfrac{\partial x_1}{\partial y}\dfrac{\partial x_2}{\partial y}+\dfrac{\partial x_1}{\partial z}\dfrac{\partial x_2}{\partial z}\\ \\ g^{13}=g^{31}=\dfrac{\partial x_1}{\partial x}\dfrac{\partial x_3}{\partial x}+\dfrac{\partial x_1}{\partial y}\dfrac{\partial x_3}{\partial y}+\dfrac{\partial x_1}{\partial z}\dfrac{\partial x_3}{\partial z}\\ \\ g^{23}=g^{32}=\dfrac{\partial x_2}{\partial x}\dfrac{\partial x_3}{\partial x}+\dfrac{\partial x_2}{\partial y}\dfrac{\partial x_3}{\partial y}+\dfrac{\partial x_2}{\partial z}\dfrac{\partial x_3}{\partial z} \end{cases} g11=(x1x)2+(x1y)2+(x1z)2g22=(x2x)2+(x2y)2+(x2z)2g33=(x3x)2+(x3y)2+(x3z)2g12=g21=x1xx2x+x1yx2y+x1zx2zg13=g31=x1xx3x+x1yx3y+x1zx3zg23=g32=x2xx3x+x2yx3y+x2zx3z g11=(xx1)2+(yx1)2+(zx1)2g22=(xx2)2+(yx2)2+(zx2)2g33=(xx3)2+(yx3)2+(zx3)2g12=g21=xx1xx2+yx1yx2+zx1zx2g13=g31=xx1xx3+yx1yx3+zx1zx3g23=g32=xx2xx3+yx2yx3+zx2zx3
说明:在笛卡尔直角坐标系中协变基矢与逆变基矢完全重合,即 e ⃗ i = g ⃗ i = g ⃗ i   ( i = 1 , 2 , 3 ) \vec{e}_i=\vec{g}_i=\vec{g}^i\ (i=1,2,3) e i=g i=g i (i=1,2,3)

评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值