求解约束优化问题的微粒群算法  

Particle Swarm Optimization for the Constrained Optimization Problem

作　　者：谭瑛[1] 高慧敏[1] 曾建潮[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：《太原重型机械学院学报》2004年第2期94-97,共4页Journal of Taiyuan Heavy Machinery Institute

摘　　要：微粒群算法(简称PSO算法)是一种新型的进化计算方法,已在许多领域得到了非常成功的应用。本文以约束优化问题为对象,首先介绍了采用罚函数法将约束优化问题化为无约束优化问题,和将约束优化问题转化为minmax问题,然后对无约束优化问题和minmax问题,采用PSO算法进行进化求解;在此基础上,以目标函数和约束满足分别为优化目标提出了一种双微粒群的PSO算法。仿真实验结果验证了方法的正确性与有效性。Particle Swarm Optimization(PSO) is a new evolutionary computation method,which has been successfully applied to many fields.Techniques for the application of SPO to the constrained optimization problems are described in the paper.The constrained optimization problem is transformed to unconstrained optimization problem and minmax problem respectively through penalty function.In order to overcome the difficulties of choosing the penalty function,a new PSO algorithm with two particle swarms are presented.the first one is for minimizing the objective function and the other for the constrained satisfaction.The simulation results show the correctness and efficiency of the methods.

关 键 词：微粒群算法 约束优化问题 罚函数 minmax问题 约束满足 PSO算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]