数值分析(Matlab): Newton - Raphson方法求解非线性方程组

前言

  本文中使用的是C++代码，本专题的其他数值方法是Matlab代码。

---

---

Newton - Raphson方法求解非线性方程组

迭代格式

已知如下方程组：
$\begin{array}{l}{f}_0=x_0+2x_1-5=0,\\ f_1=x_0\cdot x_1-2=0.\end{array}$
迭代格式为：
$\mathbf{Jd}=-\mathbf{F}$
$\begin{array}{l}{\mathbf{X}}_1={\mathbf{X}}_0+{\mathbf{d}}_0\\ {\mathbf{X}}_2={\mathbf{X}}_1+{\mathbf{d}}_1\\ \cdots \end{array}$
其中：
$\mathbf{J}=\left[\begin{array}{cc}\frac{\partial f_0}{\partial x_0}& \frac{\partial f_0}{\partial x_1}\\ \frac{\partial f_1}{\partial x_0}& \frac{\partial f_1}{\partial x_1}\end{array}\right]$
$\mathbf{d}=\left[\begin{array}{cc}\Delta x_0& \Delta x_1\end{array}\right]$${}^{\mathrm{T}}$
$\mathbf{F}=\left[\begin{array}{cc}{f}_0& f_1\end{array}\right]$${}^{\mathrm{T}}$

C++代码如下

Tip：代码使用了Eigen库

#include <iostream>
#include "..//eigen/Eigen/Eigen"

using namespace std;
using namespace Eigen;

int main()
{
	double x01 = 0.5, x02 = 1.0;			// 迭代初值
	cout << "x1 = " << x01 << endl
		<< "x2 = " << x02 << endl
		<< "Newton Raphson Method taking!" << endl;
	
	double epsilon = 1.0;					// ε表征误差容错
	VectorXd x0(2),x1(2),f(2);
	x0 << x01, x02;
	MatrixXd J(2, 2);

	while (epsilon > 1.0e-5) {
		f << x0(0) + 2 * x0(1) - 5.0,		// f0
			x0(0)* x0(1) - 2.0;				// f1
		J << 1.0, 1.0,
			x0(1), x0(0);
		x1 = J.lu().solve(-1.0 * f);		// 此时x1的含义为Δx
		epsilon = x1.norm();
		x0 += x1;
	}
	cout << x0 <<endl;

	system("pause");
	return 0;
}

输出结果：

x1 = 0.5
x2 = 1
Newton Raphson Method taking!
Vector x0:
  4 0.5