基于粒子进化的多粒子群优化算法  

Particle Swarm Optimization based on particle evolution.

作　　者：张文爱[1] 刘丽芳[1] 李孝荣[1] 

机构地区：[1]太原理工大学信息工程学院,太原030024

出　　处：《计算机工程与应用》2008年第7期51-53,共3页Computer Engineering and Applications

摘　　要：提出了一种基于粒子进化的多粒子群优化算法。该算法采用局部版的粒子群优化方法,多个粒子群彼此独立地搜索解空间,从而增强了全局搜索能力;利用重置进化粒子位置的方法使陷入局部值的粒子摆脱局部最小,从而有效地避免了"早熟"问题,提高了算法的稳定性。对3个测试函数进行了对比实验,结果表明该算法优于标准粒子群算法。The Panicle Swarms Optimization（PSO） based on panicle evolution is proposed.Location best version of PSO is adopted in the algorithm.Panicle swarms are employed to search in the solution space independently that enhances the global searching ability.The location of evolutional panicles will be reset in order to force it getting out of locally minimum.It makes the particle escaped from the premature convergence and increases the stability of algorithm.Comparative experiments on three testing functions indicate that the algorithm is better than the standard PSO.

关 键 词：粒子群算法 进化计算 群集智能 局部版粒子群算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]