结合混沌搜索的自适应混沌粒子群算法

Adaptive Chaos Particle Swarm Optimization Combined with Chaos Search

针对基于群体适应度方差的自适应混沌粒子群算法存在的局部搜索能力较弱的不足,在该算法中引入了混沌变异以及混沌搜索操作。使用An混沌映射对部分粒子进行混沌变异,对全局最优粒子进行混沌搜索,提出了一种综合考虑粒子位置、寻优空间的自适应变尺度规则。数值仿真结果表明,改进算法的收敛性、全局和局部搜索能力都有所提高,能有效避免早熟收敛。

In order to improve the local search ability of adaptive chaos particle swarm optimization based on colony fitness variance(ACPSO),chaos mutation and chaos search into ACPSO are introduced.A chaos mapping is used for chaos mutation for some particles,and chaos search is performed for the global optimal particle.The results of the numerical simulation indicates that the convergence,global searching ability and local searching ability of the presented algorithm are enhanced,the algorithm can effectively avoid trapping in local minima.