一种保证全局收敛的PSO算法  

A Guaranteed Global Convergence Particle Swarm Optimizer

作　　者：曾建潮[1] 崔志华[1] 

机构地区：[1]太原重型机械学院系统仿真与计算机应用研究所,太原030024

出　　处：《计算机研究与发展》2004年第8期1333-1338,共6页Journal of Computer Research and Development

基　　金：教育部科学技术研究重点项目 ( 2 0 40 18)

摘　　要：在对基本PSO算法分析的基础上 ,提出了一种能够保证以概率 1收敛于全局最优解的PSO算法———随机PSO算法 (stochasticPSO ,SPSO) ,并利用Solis和Wets的研究结果对其全局收敛性进行了理论分析 ,给出了两种停止进化微粒的重新产生方法A new particle swarm optimizer, called stochastic PSO, that is guaranteed to converge to the global optimization solution with probability one, is presented based on the analysis of the standard PSO. And the global convergence analysis is made using the Solis and Wets' research results, and two methods of stopping evolution particle to be regenerated are given. Finally, several examples are simulated to show that SPSO is more efficient than the standard PSO.

关 键 词：微粒群算法(PSO算法) 全局最优性 收敛性 模拟退火 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]