自适应扩展的简化粒子群优化算法  

Simplified particle swarm optimization algorithm with adaptive extended operator

作　　者：赵志刚[1] 张振文[1] 张福刚[1] 

机构地区：[1]广西大学计算机与电子信息学院,南宁530004

出　　处：《计算机工程与应用》2011年第18期45-47,共3页Computer Engineering and Applications

基　　金：国家自然科学基金(No.61063031);广西教育厅科研项目(桂教科研200626)~~

摘　　要：针对基本粒子群优化算法易于陷入局部最优的问题,提出了一种自适应扩展的简化粒子群优化算法。该算法采用去除速度项的简化算法结构,并用所有粒子个体极值的平均值代替每个粒子的个体极值,自适应动态调整加速系数。实验结果表明,算法能够有效避免早熟收敛问题,其全局收敛性能显著提高,收敛速度更快。An improved Particle Swarm Optimization（PSO） algorithm is presented based on three methods of improvement in standard PSO to avoid being trapped in local minima.The iteration formula of PSO is changed and simplified by removal of velocity parameter that is unnecessary during the course of evolution.The personal best value of each particle is replaced by the mean value of them of all particles.The acceleration coefficients are adaptively adjusted to improve the search performance of algorithm.The experimental results show that the proposed algorithm not only has great advantages of convergence property over standard PSO and some other modified PSO algorithms,but also avoids effectively being trapped in local minima.

关 键 词：粒子群优化算法 局部最优 个体极值 加速系数 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]