一种基于轮盘赌选择遗传算法的随机微粒群算法  

A Stochastic Particle Swarm Optimization Algorithm Based on the Genetic Algorithm of Roulette Wheel Selection

作　　者：夏桂梅[1] 曾建潮[1] 

机构地区：[1]太原科技大学系统仿真与计算机应用研究所,山西太原030024

出　　处：《计算机工程与科学》2007年第6期51-54,共4页Computer Engineering & Science

基　　金：教育部重点科技项目(204018)

摘　　要：以保证全局收敛的随机微粒群算法SPSO为基础,本文提出了一种改进的随机微粒群算法——GAR-SPSO。该方法是在SPSO的进化过程中,以轮盘赌选择机制下的遗传算法所产生的最优个体来代替SPSO中停止的微粒,参与下一代的群体进化。通过对五个多峰的测试函数进行仿真明:在搜索空间维数相同的情况下,GAR-SPSO收敛率及收敛速度均大大优于SPSO。Based on the stochastic particle swarm optimization algorithm that guarantees global convergence, an improved stochastic particle swarm optimization algorithm named GAR-SPSO is proposed. During the evolution of SPSO, the optimal particles produced by the genetic algorithm of roulette wheel selection substitute for the stopping particles, and take part in the evolution of the next generation. Through the experiments of five multi-modal test functions, the result of simulation proves that the speed of convergence and the rate of convergence for GAR-SPSO are better than SPSO in case of the same dimension of the search space.

关 键 词：随机微粒群算法 遗传算法 轮盘赌选择 全局优化 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]