矢量分析和张量分析

最新推荐文章于 2021-05-08 17:44:25 发布

东门之杨

最新推荐文章于 2021-05-08 17:44:25 发布

阅读量4.5k

点赞数 5

分类专栏：电动力学文章标签：电动力学

本文链接：https://blog.csdn.net/u013795675/article/details/48727609

版权

电动力学专栏收录该内容

17 篇文章 1 订阅

订阅专栏

矢量分析

矢量分析的方法参见量子力学之数学基础

以下是电动力学中常用而一般数学书中不常见的公式（俞允强《电动力学简明教程》），附带证明.

$\nabla (φ ψ) = φ \nabla ψ + ψ \nabla φ$ $\nabla(\varphi\psi)=\varphi\nabla \psi+\psi\nabla\varphi$

proof

$[\nabla (φ ψ)] i = \partial i (φ ψ) = φ \partial i ψ + ψ \partial i φ ∴ \nabla (φ ψ) = φ \nabla ψ + ψ \nabla φ$ $\begin{array} > [\nabla(\varphi\psi)]_i=\partial_i(\varphi\psi)=\varphi\partial_i \psi+\psi\partial_i\varphi\\ \therefore \nabla(\varphi\psi)=\varphi\nabla \psi+\psi\nabla\varphi \end{array}$

$\nabla \cdot (φ A) = φ \nabla \cdot A + A \cdot \nabla φ$ $\nabla\cdot(\varphi\mathbf A) = \varphi\nabla\cdot \mathbf A+\mathbf A\cdot\nabla\varphi$

proof

$\nabla \cdot (φ A) = \partial i (φ A) i = φ \partial i A i + (\partial i φ) A i = φ \nabla \cdot A + A \cdot \nabla φ$ $\begin{array} >\nabla\cdot(\varphi\mathbf A)=\partial_i(\varphi\mathbf A)_i\\ =\varphi\partial_i A_i+(\partial_i\varphi )A_i\\ =\varphi\nabla \cdot\mathbf A+\mathbf A\cdot\nabla\varphi\ \end{array}$

$\nabla \times (φ A) = φ \nabla \times A + \nabla φ \times A$ $\nabla\times(\varphi\mathbf A) = \varphi\nabla\times\mathbf A+\nabla\varphi\times\mathbf A$

proof

$[\nabla \times (φ A)] i = ε i j k \partial j (φ A) k = ε i j k [φ \partial j A k + (\partial j φ) A k] = φ ε i j k \partial j A k + ε i j k (\partial j φ) A k = [φ \nabla \times A] i + [\nabla φ \times A] i ∴ \nabla \times (φ A) = φ \nabla \times A + \nabla φ \times A$ $\begin{array} >[\nabla\times(\varphi\mathbf A) ]_i=\varepsilon_{ijk}\partial_j(\varphi\mathbf A)_k\\ =\varepsilon_{ijk}[\varphi\partial_j A_k+(\partial_j\varphi )A_k]\\ =\varphi\varepsilon_{ijk}\partial_j A_k+\varepsilon_{ijk}(\partial_j\varphi )A_k\\ =[\varphi\nabla\times\mathbf A]_i+[\nabla\varphi\times\mathbf A]_i\\ \therefore \nabla\times(\varphi\mathbf A) = \varphi\nabla\times\mathbf A+\nabla\varphi\times\mathbf A \end{array}$

$\nabla \times (\nabla \times A) = \nabla (\nabla \cdot A) - \nabla 2 A$ $\nabla\times(\nabla\times \mathbf A)=\nabla(\nabla\cdot \mathbf A)-\nabla^2\mathbf A$

proof

$[\nabla \times (\nabla \times A)] i = ε i j k \partial j (\nabla \times A) k = ε i j k \partial j [ε k l m \partial l A m] = (δ i l δ j m - δ i m δ j l) \partial j (\partial l A m) = \partial m (\partial i A m) - \partial l (\partial l A i) = \partial i (\partial m A m) - \partial l (\partial l A i) = \partial i (\nabla \cdot A) - \nabla 2 A i ∴ \nabla \times (\nabla \times A) = \nabla (\nabla \cdot A) - \nabla 2 A$ $\begin{array} >[\nabla\times(\nabla\times\mathbf A) ]_i=\varepsilon_{ijk}\partial_j(\nabla\times\mathbf A)_k\\ =\varepsilon_{ijk}\partial_j[\varepsilon_{klm}\partial_lA_m]\\ =(\delta_{il}\delta_{jm}-\delta_{im}\delta_{jl})\partial_j(\partial_l A_m)\\ =\partial_m(\partial_i A_m)-\partial_l(\partial_l A_i)\\ =\partial_i(\partial_m A_m)-\partial_l(\partial_l A_i)\\ =\partial_i(\nabla\cdot\mathbf A)-\nabla^2 A_i\\ \therefore\nabla\times(\nabla\times \mathbf A)=\nabla(\nabla\cdot \mathbf A)-\nabla^2\mathbf A \end{array}$

(\nabla \times A) \times A = A \cdot \nabla A - 1 2 \nabla A 2

$(\nabla\times \mathbf A)\times \mathbf A=\mathbf A\cdot\nabla \mathbf A-\dfrac12\nabla\mathbf A^2$

proof

$[(\nabla \times A) \times A] i = ε i j k (\nabla \times A) j A k = ε i j k ε j l m (\partial l A m) A k = ε k i j ε j l m (\partial l A m) A k = (δ k l δ i m - δ k m δ i l) (\partial l A m) A k = (\partial k A i) A k - (\partial i A k) A k = (A \cdot \nabla) A i - 1 2 \partial i (A 2 k) ∴ (\nabla \times A) \times A = A \cdot \nabla A - 1 2 \nabla A 2$ $\begin{array} >[(\nabla\times \mathbf A)\times \mathbf A ]_i=\varepsilon_{ijk}(\nabla\times\mathbf A)_jA_k\\ =\varepsilon_{ijk}\varepsilon_{jlm}(\partial_lA_m)A_k\\ =\varepsilon_{kij}\varepsilon_{jlm}(\partial_lA_m)A_k\\ =(\delta_{kl}\delta_{im}-\delta_{km}\delta_{il})(\partial_lA_m)A_k\\ =(\partial_kA_i)A_k-(\partial_iA_k)A_k\\ =(\mathbf A\cdot\nabla)A_i-\dfrac12\partial_i(A_k^2)\\ \therefore(\nabla\times \mathbf A)\times \mathbf A=\mathbf A\cdot\nabla \mathbf A-\dfrac12\nabla\mathbf A^2 \end{array}$

$\nabla \times (A \times B) = A (\nabla \cdot B) + (B \cdot \nabla) A - B (\nabla \cdot A) - (A \cdot \nabla) B$ $\nabla\times (\mathbf A\times \mathbf B)=\bf A(\nabla\cdot B)+(B\cdot\nabla)A-B(\nabla\cdot A)-(A\cdot \nabla)B$

proof

$[\nabla \times (A \times B)] i = ε i j k \partial j (A \times B) k = ε i j k \partial j ε k l m (A l B m) = (δ i l δ j m - δ i m δ j l) [(\partial j A l) B m + A l (\partial j B m)] = [(\partial j A i) B j + A i (\partial j B j)] - [(\partial j A j) B i + A j (\partial j B i)] = (B \cdot \nabla) A i + A i (\nabla \cdot B) - (\nabla \cdot A) B i - (A \cdot \nabla) B i ∴ \nabla \times (A \times B) = A (\nabla \cdot B) + (B \cdot \nabla) A - B (\nabla \cdot A) - (A \cdot \nabla) B$ $\begin{array}& [\nabla\times (\mathbf A\times \mathbf B)]_i=\varepsilon_{ijk}\partial_j(\mathbf A\times \mathbf B)_k\\ =\varepsilon_{ijk}\partial_j\varepsilon_{klm}(A_lB_m)\\ =(\delta_{il}\delta_{jm}-\delta_{im}\delta_{jl})[(\partial_j A_l)B_m+A_l(\partial_j B_m)]\\ =[(\partial_j A_i)B_j+A_i(\partial_jB_j)]-[(\partial_j A_j)B_i+A_j(\partial_j B_i)]\\ =(\mathbf B\cdot\nabla) A_i+A_i(\nabla\cdot\mathbf B)-(\nabla\cdot \mathbf A)B_i-(\mathbf A\cdot\nabla) B_i\\ \therefore\nabla\times (\mathbf A\times \mathbf B)=\bf A(\nabla\cdot B)+(B\cdot\nabla)A-B(\nabla\cdot A)-(A\cdot \nabla)B \end{array}$

$\nabla (A \cdot B) = (A \cdot \nabla) B + (B \cdot \nabla) A + A \times (\nabla \times B) + B \times (\nabla \times A)$ $\nabla(\mathbf A\cdot\mathbf B)=\bf(A\cdot \nabla)B+(B\cdot\nabla)A+A\times(\nabla\times B)+B\times(\nabla\times A)$

proof

$[\nabla (A \cdot B)] i = \partial i (A j B j) = (\partial i A j) B j + A j (\partial i B j) [A \times (\nabla \times B)] i = ε i j k A j (\nabla \times B) k = ε i j k ε k l m A j (\partial l B m) = (δ i l δ j m - δ i m δ j l) A j (\partial l B m) = A j (\partial i B j) - A j (\partial j B i) = A j (\partial i B j) - (A \cdot \nabla) B i Similarly ， [B \times (\nabla \times A)] i = B j (\partial i A j) - (B \cdot \nabla) A i ∴ [(A \cdot \nabla) B + (B \cdot \nabla) A + A \times (\nabla \times B) + B \times (\nabla \times A)] i = (A \cdot \nabla) B i + (B \cdot \nabla) A i + [A j (\partial i B j) - (A \cdot \nabla) B i] + [B j (\partial i A j) - (B \cdot \nabla) A i] = (\partial i A j) B j + A j (\partial i B j) ∴ \nabla (A \cdot B) = (A \cdot \nabla) B + (B \cdot \nabla) A + A \times (\nabla \times B) + B \times (\nabla \times A)$ $\begin{array} > [\nabla(\mathbf A\cdot\mathbf B)]_i=\partial_i(A_jB_j) \\=(\partial_iA_j)B_j+A_j(\partial_iB_j)\\ [\mathbf A\times(\nabla\times \mathbf B)]_i=\varepsilon_{ijk}A_j(\nabla\times \mathbf B)_k\\ =\varepsilon_{ijk}\varepsilon_{klm}A_j(\partial_lB_m)\\ =(\delta_{il}\delta_{jm}-\delta_{im}\delta_{jl})A_j(\partial_lB_m)\\ =A_j(\partial_iB_j)-A_j(\partial_jB_i)\\ =A_j(\partial_iB_j)-(\mathbf A\cdot \nabla)B_i\\ \text{Similarly}，[\mathbf B\times(\nabla\times \mathbf A)]_i=B_j(\partial_iA_j)-(\mathbf B\cdot \nabla)A_i\\ \therefore[\bf(A\cdot \nabla)B+(B\cdot\nabla)A+A\times(\nabla\times B)+B\times(\nabla\times A)]_i \\ =(\mathbf A\cdot \nabla)B_i+(\mathbf B\cdot \nabla)A_i+[A_j(\partial_iB_j)-(\mathbf A\cdot \nabla)B_i]+[B_j(\partial_iA_j)-(\mathbf B\cdot \nabla)A_i]\\ =(\partial_iA_j)B_j+A_j(\partial_iB_j)\\ \therefore \nabla(\mathbf A\cdot\mathbf B)=\bf(A\cdot \nabla)B+(B\cdot\nabla)A+A\times(\nabla\times B)+B\times(\nabla\times A)\\ \end{array}$

$\nabla r = r r = e (r)$ $\nabla r=\dfrac{\mathbf r}r=\mathbf e(\mathbf r)$

proof

$(\nabla r) i = \partial i x j x j - - - - \sqrt = \partial i ( x 2 j ) 2 x j x j - - - - \sqrt = 2 x i 2 x j x j - - - - \sqrt = x i r ∴ \nabla r = r r = e (r)$ $\begin{array} >(\nabla r)_i=\partial_i\sqrt{x_jx_j} \\=\dfrac{\partial_i(x_j^2)}{2\sqrt{x_jx_j}} \\=\dfrac{2x_i}{2\sqrt{x_jx_j}} \\=\dfrac{x_i}{r} \\\therefore \nabla r=\dfrac{\mathbf r}r=\mathbf e(\mathbf r) \end{array}$

$\nabla f (r) = d f d r e (r)$ $\nabla f(r)=\dfrac{\mathrm df}{\mathrm dr}\mathbf e(\mathbf r)$

proof

$[\nabla f (r)] i = \partial i f (r) = d f d r \partial i r = d f d r e (r)$ $\begin{array} >[\nabla f(r)]_i=\partial_if(r) \\=\dfrac{\mathrm df}{\mathrm dr}\partial_i r \\=\dfrac{\mathrm df}{\mathrm dr}\mathbf e(\mathbf r) \end{array}$

$\nabla 2 1 r = - 4 π δ (r)$ $\nabla^2 \dfrac1r=-4\pi\delta(r)$

when r \neq 0, \partial 2 i 1 r = \partial i (\partial i 1 x j x j - - - - \sqrt) = \partial i ⎛ ⎝ - \partial i ( x j x j ) 2 ( x j x j - - - - \sqrt ) 3 ⎞ ⎠ = \partial i ⎛ ⎝ - x i ( x j x j - - - - \sqrt ) 3 ⎞ ⎠ = - 3 ( x j x j - - - - \sqrt ) 3 - x i \partial i [ ( x j x j - - - - \sqrt ) 3 ] ( x j x j - - - - \sqrt ) 6 = - 3 ( x j x j - - - - \sqrt ) 3 - 3 ( x j x j - - - - \sqrt ) 3 ( x j x j - - - - \sqrt ) 6 = 0 when r=0 \int V \nabla 2 1 r d τ = \int V \nabla \cdot (\nabla 1 r) d τ = \int \partial V (\nabla 1 r) \cdot d σ = \int \partial V [- 1 r 2 e (r)] \cdot d σ = - 4 π ∴ \nabla 2 1 r = - 4 π δ (r)

$\begin{array} &\text{when r}\neq0, \\\partial_i^2 \dfrac1r=\partial_i\left(\partial_i \dfrac1{\sqrt{x_jx_j}}\right) \\=\partial_i\left(-\dfrac{\partial_i(x_jx_j)} {2(\sqrt{x_jx_j})^3}\right) \\= \partial_i\left(-\dfrac{x_i} {(\sqrt{x_jx_j})^3}\right) \\=-\dfrac{3(\sqrt{x_jx_j})^3-x_i\partial_i[(\sqrt{x_jx_j})^3]} {(\sqrt{x_jx_j})^6} \\=-\dfrac{3(\sqrt{x_jx_j})^3-3(\sqrt{x_jx_j})^3} {(\sqrt{x_jx_j})^6} \\=0\\ \text{when r=0} \\\int_V\nabla^2 \dfrac1r\mathrm d\tau\\= \int_V\nabla\cdot (\nabla\dfrac1r)\mathrm d\tau\\= \int_{\partial V}(\nabla\dfrac1r)\cdot\mathrm d\sigma \\=\int_{\partial V}[-\dfrac1{r^2}\mathbf e(\mathbf r)]\cdot\mathrm d\sigma \\=-4\pi \\\therefore \nabla^2 \dfrac1r=-4\pi\delta(r) \end{array}$

张量分析

物理量只用一个数描述，且其值在坐标架转动下保持不变，称为标量或者零阶张量，即；

$T ¯ ¯ ¯ = T$ $\overline{T}=T$
物理量用三个数描述，且其值在坐标架转动时的变化规律和点的坐标变换一样，称为矢量或一阶张量，即

$T ¯ ¯ ¯ i = R i j T j$ $\overline{T}_i=R_{ij}T_j$
物理量用九个数描述，且其值在坐标架转动时的变化规律满足

$T ¯ ¯ ¯ i j = R i m R j n T j$ $\overline{T}_{ij}=R_{im}R_{jn}T_j$
称为三阶张量.

东门之杨

关注

5
点赞
踩
12

收藏

觉得还不错? 一键收藏
0
评论
矢量分析和张量分析

矢量分析矢量分析的方法参见量子力学之数学基础以下是电动力学中常用而一般数学书中不常见的公式（俞允强《电动力学简明教程》），附带证明.$$\nabla(\varphi\psi)=\varphi\nabla \psi+\psi\nabla\varphi$$ proof \begin{array}> [\nabla(\varphi\psi)]_i=\partial_i(\varphi\psi)
复制链接

扫一扫