基于遗传算法的非线性方程组求解

针对目前求解非线性方程组所采用的牛顿法及其变形算法存在的运算量大、求解速度慢的问题，提出了一个求解非线性方程组近似解的通用遗传算法。该算法主要采用求解目标函数极小值的思想，并结合遗传算法并行搜索的特点，通过选择和设置适当的父体选择策略、杂交算子、变异算子等参数，使算法取得了较高的收敛速度和精度。实验结果表明，该方法明显优于传统方法，并具有运算速度快、精度高、通用性好的特点。

Approach for Solving Nonlinear Equation Group Based on Genetic Algorithm

For the existing problems of much computing and slow speed in solving nonlinear equation group by Newton iteration method and its transfigurations, a general genetic algorithm of seeking approximate solutions of nonlinear equation group is proposed. Having mainly adopted the idea of solving the minimum value of objective function and combined with the parallel search merit of genetic algorithm, the algorithm obtains fast convergence rate and high precision by selecting and setting appropriate parameters such as parent selection strategy, crossover operator, mutation operator, etc. The experimental results show that the method obviously is superior to conventional methods, and it has characteristics of rapid computation, high precision and well universality.