连带 Legendre 函数¶
回顾
- 波动方程
\[u_{tt} - a^2u_{xx} = f(x, t)\]
- 输运问题
\[u_t - a^2 u_{xx} = f(x, t)\]
- 稳定场分布
\[\nabla^2 u = f(\vec{r})\]解的适定性:
- 存在性
- 唯一性
- 同一时刻同一位置的物理量应该是唯一的
- 稳定性
- 允许个别点的矛盾/挖掉个别点不影响整体结果
- 一维波动方程
\[u_{tt} - a^2 u_{xx} = f(x, t), \quad u|_{x=a} = u|_{x=b} = 0\]
- 齐次化,分离变量法:\(u(x, t) = X(x)T(t)\),得到
\[\frac{T''(t)}{a^2 T(t)} = \frac{X''(x)}{X(x)} = -\lambda\]求得本征函数 \(X(x) = A\sin \frac{(x-a)n\pi}{b-a}\).
- 广义傅里叶级数展开
\[\left(T_n''(t) - a^2 \left(\frac{n\pi}{b-a}\right)^2\right)\sin \frac{(x-a)n\pi}{b-a} = f(x, t)\]
母函数
\[ \frac{1}{\sqrt{R^2 + r^2 - 2Rr\cos\theta}} = \begin{cases} \displaystyle \sum_{l=0}^{\infty} \frac{r^l}{R^{l+1}} P_l(\cos\theta), & r < R \\[2em] \displaystyle \sum_{l=0}^{\infty} \frac{R^l}{r^{l+1}} P_l(\cos\theta), & r > R \end{cases} \]
球坐标 Laplace 方程
\[\nabla^2 u = 0, \quad u = u(r, \theta, \varphi) = R(r)\Theta(\theta)\Phi(\varphi)\]
\[\frac{1}{r^2} \frac{\partial}{\partial r} \left(r^2 \frac{\partial R}{\partial r}\right) + \frac{1}{r^2\sin\theta} \frac{\partial}{\partial \theta} \left(\sin\theta \frac{\partial \Theta}{\partial \theta}\right) + \frac{1}{r^2\sin^2\theta} \frac{\partial^2 \Phi}{\partial \varphi^2} = 0\]
\[\Phi''(\varphi) + m^2 \Phi(\varphi) = 0 \Rightarrow \Phi(\varphi) = A\cos m\varphi + B\sin m\varphi,\, m \in \mathbb{Z}\]
\[\sin \theta \frac{d}{d\theta} \left(\sin\theta \frac{d\Theta}{d\theta}\right) + \left[l(l+1) - \frac{m^2}{\sin^2\theta}\right]\Theta = 0\]
作变换 \(x = \cos\theta\),得到
\[(1-x^2) \Theta''(x) - 2x\Theta'(x) + \left[l(l+1) - \frac{m^2}{1-x^2}\right]\Theta(x) = 0\]
如果直接代入 \(\Theta(x) = \sum_{k=0}^{\infty} a_k x^k\),会得到 \(a_{k+2}, a_k, a_{k-2}\) 三项之间的关系,极为复杂!
由于 \(1-x^2\) 反复出现,尝试令 \(\Theta(x) = (1-x^2)^A y(x)\),为了消去 \(y(x)\) 前面含 \(x\) 的讨厌的项(这会使递推关系复杂),有 \(4A^2 x^2 - m^2 = m^2(1-x^2)\),即 \(A = \pm \frac{m}{2}\). 于是
\[(1-x^2)y''(x) - 2(m+1)xy'(x) + \left[l(l+1) - m(m+1)\right]y(x) = 0\]
将 \(m=0\) 的 Legendre 方程求 \(m\) 次导:
\[\left[(1-x^2)P_l''(x) - 2xP_l'(x) + l(l+1)P_l(x)\right]^{[m]} = 0\]
得到
\[(1-x^2)P_l^{[m+2]}(x) - 2(m+1)xP_l^{[m+1]}(x) + \left[l(l+1) - m(m+1)\right]P_l^{[m]}(x) = 0\]
对比得到 \(y(x) = P_l^{[m]}(x)\),收敛半径为 \(|x| < 1\). 故
\[\Theta(x) = (1-x^2)^{\frac{m}{2}} P_l^{[m]}(x) \equiv P_l^{m}(x)\]
称为连带 Legendre 函数。因为 \(P_l(x)\) 最高次幂为 \(x^l\),所以求导次数 \(|m| \leq l\)(不然就变成 \(0\) 了)。
- 微分表示
\[\Theta(x) = P_l^m(x) = \frac{1}{2^l l!} (1-x^2)^{\frac{m}{2}} \frac{d^{l+m}}{dx^{l+m}}(x^2-1)^l\]
\(m\) 为负时,
\[\tilde{\Theta}(x) = P_l^{-m}(x) = (1-x^2)^{-\frac{m}{2}} \frac{1}{2^l l!} \frac{d^{l-m}}{dx^{l-m}}(x^2-1)^l\]
因为不满足自然边界条件的解扔掉了一个,所以 \(P_l^m(x)\) 和 \(P_l^{-m}(x)\) 应是线性相关的,两者只差一个常数因子。
\[\frac{P_l^m(x)}{P_l^{-m}(x)} = \frac{(1-x^2)^m \frac{d^{l+m}}{dx^{l+m}}(x^2-1)^l}{(1-x^2)^{-m} \frac{d^{l-m}}{dx^{l-m}}(x^2-1)^l} = (-1)^m \frac{(l+m)!}{(l-m)!}\]
这个结果在两个勒让德函数相乘后计算积分时(例如下面的证明正交性)比较好用,因为 \(P_l^m(x)P_k^m(x)\) 相乘会出来一个 \((1-x^2)^m\),会很麻烦,可以把其中一个换成 \(P_l^{-m}(x)\),这样就可以消去。
- 积分公式
\[f(x) = \frac{1}{2\pi \mathrm{i}} \oint \frac{f(z)}{z-x} dz\]
- 正交性
\[ \begin{aligned} \int_{-1}^{1} P_l^m(x) P_k^m(x) dx &= (-1)^m \frac{(k+m)!}{(k-m)!} \int_{-1}^{1} P_l^m(x) P_k^{-m}(x) dx \\ &= (-1)^m \frac{(k+m)!}{(k-m)!} \int_{-1}^{1} P_l^{[m]}(x) P_k^{[m]}(x) dx \\ &= (-1)^m \frac{(k+m)!}{(k-m)!} \frac{1}{2^l l!} \frac{1}{2^k k!} \int_{-1}^{1} \frac{d^{l+m}}{dx^{l+m}}(x^2-1)^l \frac{d^{k-m}}{dx^{k-m}}(x^2-1)^k dx \\ &= (-1)^m \frac{(k+m)!}{(k-m)!} \frac{1}{2^l l!} \frac{1}{2^k k!} \int_{-1}^{1} d\frac{d^{l+m-1}}{dx^{l+m-1}}(x^2-1)^l \frac{d^{k-m}}{dx^{k-m}}(x^2-1)^k \\ &= (-1)^m \frac{(k+m)!}{(k-m)!} \frac{1}{2^l l!} \frac{1}{2^k k!} \left(\left. \frac{d^{l+m-1}}{dx^{l+m-1}}(x^2-1)^l \frac{d^{k-m}}{dx^{k-m}}(x^2-1)^k \right|_{-1}^{1} - \int_{-1}^{1} \frac{d^{l+m-1}}{dx^{l+m-1}}(x^2-1)^l \frac{d^{k-m+1}}{dx^{k-m+1}}(x^2-1)^k dx \right) \\ &= (-1)^{m+1} \frac{(k+m)!}{(k-m)!} \frac{1}{2^l l!} \frac{1}{2^k k!} \int_{-1}^{1} \frac{d^{l+k}}{dx^{l+k}}(x^2-1)^{l+k} dx \\ &= (-1)^{m+1} \frac{(k+m)!}{(k-m)!} \frac{1}{2^l l!} \frac{1}{2^k k!} \int_{-1}^{1} \frac{d^l}{dx^l} (x^2-1)^l \frac{d^k}{dx^k} (x^2-1)^k dx \\ &= (-1)^{m+1} \frac{(k+m)!}{(k-m)!} \frac{1}{2^l l!} \frac{1}{2^k k!} \int_{-1}^{1} P_l(x) P_k(x) dx \\ &= (-1)^{m+1} \frac{(k+m)!}{(k-m)!} \frac{1}{2^l l!} \frac{1}{2^k k!} \delta_{lk} \end{aligned} \]