基于共轭梯度法和Tent映射的拟态物理学算法

摘要

An improved artificial physics algorithm which is combined conjugate gradient method with chaotic perturbations is proposed to solve the problem that the improved artificial physics algorithm has poor search accuracy in later period and it is easy to fall into local optimum.This algorithm adopted conjugate gradient method,a high accuracy analytical algorithm,to search locally when it is difficult for artificial physics algorithm to search accurately in later period,replacing artificial physics algorithm.It also introduced the Chaotic Perturbations to the whole algorithm,to a certain extent,to prevent the algorithm convergent too early.Tests show that the algorithm could jump out from the local optimal solution and had better solution in precision,stability and speed.So this algorithm is suitable for the high dimension complex functions optimization.%针对拟态物理学算法优化算法后期搜索精度差、陷入局部最优的问题,提出一种融合共轭梯度法和混沌扰动的改进拟态物理学算法.该算法是在拟态物理学算法后期难以精细搜寻时,采用高精度解析算法——共轭梯度法替代拟态物理学算法进行局部搜寻;在整个算法中加入混沌扰动,避免算法“早熟”.仿真结果表明该算法收敛速度快、精度高,在跳出局部最优解上有明显优势.因此该算法适应于高维的复杂函数的寻优.