几种改进粒子群优化算法的性能分析与研究.pdfVIP

中国科技论文在线 几种改进粒子群优化算法的性能分析与 研究 胡建明，马开良，牛明飞** 5 （兰州大学数学与统计学院，兰州 730000 ） 摘要：粒子群优化算法（PSO ）是一种群体智能优化算法，其思想来源于人工生命和演化计 算理论。PSO 算法通过粒子追随自己找到的最好解和整个种群的最好解来完成优化，其优 势在于简单容易而功能强大，其缺点在于容易过早成熟和陷入局部最优。本文对基本 PSO 算法（BPSO ）和其几种改进PSO 方法做了一些探讨研究，几种改进方法分别是是具有自适 10 应随机惯性权重PSO 算法（WPSO ）、量子行为PSO 算法（QPSO ）、混沌PSO 算法（CPSO ）。 用这四种优化算法对 6 个常用测试函数进行优化计算, 仿真结果表明混沌粒子群优化算法 （CPSO ）比其他微粒群优化算法具有更好的优化性能。 关键词：优化；粒子群优化算法；自适应随机惯性权重；量子行为；混沌搜索 中图分类号：TP301 15 Performance analyzing and researching of several improved algorithm HU Jianming, MA Kailiang, NIU Mingfei (School of Mathematics and Statistics, Lanzhou University, LanZhou 730000) 20 Abstract: Particle Swarm Optimization (PSO) is an evolutionary swarm optimization technique, which comes from the artificial life and evolutionary computation theory. PSO algorithm accomplish the optimization problems through the method that each particle keeps track of its coordinates in the problem space which are associated with the best solution (fitness) it has achieved so far and at the same time another “best” value that is tracked by the global version of 25 the particle swarm optimizer is the overall best value, and its location, obtained so far by any particle in the population. The advantages of PSO are effective and powerful, while the drawbacks are that easy to prematur