基于Morlet小波变异的粒子群优化算法  

A Particle Swarm Optimization Algorithm Based on Morlet Wavelet Mutation

作　　者：张超[1] ZHANG Chao(Department of Computer Information,Suzhou Vocational and Technological College,Suzhou 234101,Anhui,China)

机构地区：[1]宿州职业技术学院计算机信息系,安徽宿州234101

出　　处：《江汉大学学报（自然科学版）》2018年第2期109-119,共11页Journal of Jianghan University：Natural Science Edition

基　　金：安徽省高校省级自然科学基金重点项目(KJ2016A781;KJ2016A778);安徽省高校省级质量工程项目(2015jyxm512)

摘　　要：针对粒子群优化算法易陷入局部极值,收敛精度不高的缺陷,提出一种基于Morlet小波变异的改进算法。改进算法对组成每代全局极值的各维度实施小波扰动,并将扰动结果作为以一定概率被选中粒子的新位置,充分利用全局极值的优势信息引导粒子快速向最优解靠近,通过小波函数的微调特征帮助粒子跳出局部极值。在12个经典测试函数上的仿真实验结果表明,改进算法的寻优性能较SPSO、CLPSO、DEOPSO、HPSOWM算法有显著提高,适合于求解函数优化问题。A particle swarm optimization algorithm based on Morlet wavelet mutation was presented to overcome the problems of low convergence precision and easily falling into local extremum.Morlet muta.tion operation was implemented for each dimension of global extremum,the mutation results were used as new positions of particles,which was selected in certain probability.This strategy made full use of the advantage information of global extremum to guide the particle to approach the optimal solution quickly.At the same time,the fine tuning feature of wavelet function helped the particle jumping out of the local extremum.The simulation experiments on 12 classical test functions showed that the improved algorithm had better performance than SPSO,CLPSO,DEOPSO and HPSOWM algorithms and was suitable for solving function optimization problems.

关 键 词：粒子群优化算法 MORLET小波 收敛速度 收敛精度 时间复杂度 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]