带Levy变异的约束优化PSO算法  

Constrained Optimization PSO Algorithm with Levy Mutation

作　　者：刘淳安[1] 

机构地区：[1]宝鸡文理学院数学系,陕西宝鸡721013

出　　处：《西华大学学报（自然科学版）》2008年第2期72-75,共4页Journal of Xihua University:Natural Science Edition

基　　金：陕西省自然科学基础研究计划项目(No.2006A12);陕西省教育厅科学研究计划项目(No.07JK180);宝鸡文理学院重点科研计划项目(No.ZK0619)

摘　　要：提出了一种解约束优化问题的新PSO算法(LCPSO)。该方法引入了Levy变异策略,使算法LCPSO能有效克服标准PSO算法易陷入局部最优的缺陷。为更好求解约束边界附近的全局最优解,算法在选择下一代个体时保持群体中不可行解的一定比例,这样,不但能有效增加群体的多样性,而且避免了传统的过度惩罚,使群体向最优解更好、更快的逼近。数值试验表明该算法对约束优化问题求解是非常有效的。A new PSO algorithm (LCPSO) for solving the constrained optimization problems is presented in this paper. The new approach does not require the use of a penalty function, it uses a Levy mutation operator to overcome the defaults of plunging into the local optimal solution by simple PSO. In order to obtain the global optimal solutions, which often locate in the boundary of the constrained region, a new selection operator is introduced based on the constrained conditions of the optimization problem. The new LCPSO can keep the ratio of infeasible solutions in the swarm when selecting the next generation swarm by using the new selection operator. As a result, it can not only increase the diversity of swarm but also avoid the defects of over-penalization and make the swarm approach to the optimal solutions. The numerical experiments show that the proposed algorithm is effective in dealing with the constrained optimization problems.

关 键 词：约束优化 PSO算法 Levy变异 种群多样性 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] O221[自动化与计算机技术—控制科学与工程]