Skip to content

Commit

Permalink
fix small bugs
Browse files Browse the repository at this point in the history
  • Loading branch information
CJL-sysu committed Jun 9, 2024
1 parent b67e626 commit 988cd3a
Show file tree
Hide file tree
Showing 4 changed files with 5 additions and 5 deletions.
2 changes: 1 addition & 1 deletion 2/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -77,7 +77,7 @@ $f[x_0, x_1, \dots, x_n]=\displaystyle \frac{f^{(n)}(\xi)}{n!},\xi\in[a, b] $

**分段线性插值**: $I_h(x)=\displaystyle \frac{x-x_{k+1}}{x_k-x_{k+1}}f(x_k)+\frac{x-x_k}{x_{k+1}-x_k}f(x_{k+1}) $, 其中 $x_k\leqslant x\leqslant x_{k+1}, k=0, 1,\dots, n-1 $

**三次样条插值函数**: $S_i(x)=a_i(x-x_i)^3+b_i(x-x_i)^2+c_I(x-x_i)+d_i, x\in [x_i, x_{i+1}] $
**三次样条插值函数**: $S_i(x)=a_i(x-x_i)^3+b_i(x-x_i)^2+c_i(x-x_i)+d_i, x\in [x_i, x_{i+1}] $

记 $h$ 为小区间长度, $M_i=S''(x_i), y_i=f(x_i)$ ,在区间 $[x_i, x_{i+1}]$ 上的 $S(x)$ 为

Expand Down
2 changes: 1 addition & 1 deletion 3/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -142,6 +142,6 @@ $ P_0(t)=1, \quad P_1(t)=2t, \quad P_{k+1}(t)=2tP_k(t)-2kP_{k-1}(t) $

如果拟合的时候,用的是正交多项式 $\{p_i(x)\}$ ,可以直接写出最佳平方逼近函数 $S^*(x)=\displaystyle \sum_{k=0}^{n}\frac{ (f, p_k)}{(p_k, p_k)}p_k(x)$,注意此时的积分区域为定义域,非定义域需做出相应变换

对于一般的积分,有: $\displaystyle \int_a^b P(s)ds= \frac{b-a}{2}\int_{-1}^{1} P(t) dt,\ s=\frac{b-a}{2}+\frac{b+a}{2}$
对于一般的积分,有: $\displaystyle \int_a^b P(s)ds= \frac{b-a}{2}\int_{-1}^{1} P(t) dt,\ s=\frac{b-a}{2}t+\frac{b+a}{2}$

对于 Legendre 多项式,有: $\displaystyle \int_{-1}^1 P_j(t)P_k(t)dt =\frac{2}{2k+1},\ j= k$
2 changes: 1 addition & 1 deletion 5/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@ $$

若 $a_{kk}^{(k)}\neq 0$ ,则 $\displaystyle x_n=({b_k^{(k)}-\sum_{j=k+1}^{n}a_{kj}^{(k)}x_j})/({a_{kk}^{(k)}})$

**LU 分解**:高斯消元法的每一步初等变换相当于一个初等矩阵左乘原矩阵,最终得到一个上三角阵 $\mathbf{L_{n-1}\dots L_2L_1A=U}$
**LU 分解**(Doolittle)高斯消元法的每一步初等变换相当于一个初等矩阵左乘原矩阵,最终得到一个上三角阵 $\mathbf{L_{n-1}\dots L_2L_1A=U}$

所以 $\mathbf{A=L_1^{-1}L_2^{-1}\dots L_{n-1}^{-1}U=LU}$ ,其中 $\mathbf{L}$ 是下三角阵:

Expand Down
4 changes: 2 additions & 2 deletions 6/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -38,7 +38,7 @@ $$

**逐次超松驰迭代法(SOR)**

引入松弛因子 $\omega, \omega>0$ 采用矩阵分裂: $\displaystyle \mathbf{A=M-N}=\frac 1 \omega(\mathbf{D}-\omega\mathbf{L})-\frac 1 \omega[(1-\omega)\mathbf{D}+\omega\mathbf{U}]$
引入松弛因子 $\omega, 0<\omega<2$ 采用矩阵分裂: $\displaystyle \mathbf{A=M-N}=\frac 1 \omega(\mathbf{D}-\omega\mathbf{L})-\frac 1 \omega[(1-\omega)\mathbf{D}+\omega\mathbf{U}]$

则 $\mathbf{Dx^{(k+1)}=Dx^{(k)}}+\omega\mathbf{(b+Lx^{(k+1)}+Ux^{(k)}-Dx^{(k)})}$

Expand All @@ -52,7 +52,7 @@ $$

**严格对角占优矩阵**

满足条件 $\sum_{j=1}^n\left|a_{ij}\right|<\left|a_{ii}\right|\quad,\quad i=1,2,\cdots,n$ ,则称矩阵A是严格对角占优矩阵。
满足条件 $\sum_{j=1,j\ne i}^n\left|a_{ij}\right|<\left|a_{ii}\right|\quad,\quad i=1,2,\cdots,n$ ,则称矩阵A是严格对角占优矩阵。

雅可比迭代法和高斯-赛格尔迭代法都收敛

Expand Down

0 comments on commit 988cd3a

Please sign in to comment.