基于自适应惯性权重的混沌粒子群算法  

Chaos particle swarm optimization based on the adaptive inertia weight

作　　者：周燕[1,2] 刘培玉[1,2] 赵静[1,2] 王乾龙[1,2] 

机构地区：[1]山东师范大学信息科学与工程学院,山东济南250014 [2]山东省分布式计算机软件新技术重点实验室,山东济南250014

出　　处：《山东大学学报（理学版）》2012年第3期27-32,共6页Journal of Shandong University(Natural Science)

基　　金：国家自然科学基金资助项目(60873247);山东省高新自主创新专项工程项目(2008ZZ28);山东省自然科学基金重点项目(ZR2009GZ007)

摘　　要：针对粒子群优化(particle swarm optimization,PSO)算法易陷入早熟的缺陷,提出了一种基于自适应惯性权重的混沌粒子群算法。首先利用立方映射产生的混沌序列对粒子位置进行初始化,为全局搜索的多样性奠定基础;然后采用自适应惯性权重优化策略,提高收敛速度;最后如果判断算法陷入早熟,则对算法进行混沌扰动,使其跳出局部最优。仿真实验结果表明,改进算法的收敛速度及收敛精度都有明显提高,能有效地避免早熟。Aiming at the premature convergence problem which the particle swarm optimization algorithm suffers from,a chaos particle swarm optimization based on adaptive inertia weight is proposed. Firstly, chaotic sequence generated by cube map is used to initiate individual position, which strengthens the diversity of global searching. Secondly, adaptive inertia weight is adopted to improve the convergence rate. Furthermore, chaos perturbation is utilized to avoid the prema- ture convergence. The results of the simulation experiment show that the convergence rate and the precision of the im- proved algorithm are obviously enhanced, and the algorithm can effectively avoid the premature convergence problem.

关 键 词：粒子群算法 立方映射 自适应惯性权重 混沌扰动 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]