一种新的混合粒子群方法  

New hybrid particle swarm optimization

作　　者：管月智[1] 葛洪伟[1] 

机构地区：[1]江南大学信息工程学院,江苏无锡214122

出　　处：《计算机工程与应用》2011年第23期45-47,50,共4页Computer Engineering and Applications

摘　　要：利用混沌搜索和变异机制克服种群易停滞且易陷入局部最优点的不足。当种群出现停滞时先用混沌搜索更优点,当搜索到的点不满足变异精度要求时再进行变异。发现混沌搜索能使种群在出现停滞时持续寻优,而变异机制则能够有效地帮助种群在陷入局部最优点时跳出该点。结果表明该方法的全局寻优能力较强。By introducing chaotic search and mutation mechanism,population can overcome the shortcomings of stagnation and trapping into local optimums.When the population is stagnation,chaotic search is used to find a better point firstly.If this point does not meet the precision of mutation,then mutation mechanism will make effort.Chaotic search can continuously optimize when the population is stagnation,mutation mechanism helps the population break away from local optimum when it has trapped into the local optimum.Experimental results show that the new algorithm is relatively obvious for the ability of global optimization.

关 键 词：混沌 变异 粒子群优化 变异机制 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]