基于自适应变异的混沌粒子群优化算法  

Chaotic particle swarm optimization algorithm with adaptive mutation

作　　者：李建美[1] 高兴宝[1] 

机构地区：[1]陕西师范大学数学与信息科学学院,西安710062

出　　处：《计算机工程与应用》2016年第10期44-49,共6页Computer Engineering and Applications

基　　金：国家自然科学基金(No.61273311);中央高校基本科研业务费专项基金资助(No.GK201001002)

摘　　要：粒子群优化算法参数少,寻优速度快,但其寻优效率低且在寻优后期易早熟收敛。为改善其寻优性能,在标准粒子群优化算法中,通过引入混沌映射和自适应变异策略,提出具有自适应变异的混沌粒子群优化(ACPSO)算法,以增强种群的全局寻优性能和局部寻优效率。六个基准测试函数的仿真结果表明,ACPSO算法比已有的五个算法具有更好的寻优能力。Basic particle swarm optimization algorithm has fewer parameters and fast optimization speed, but it suffers some drawbacks such as low optimization efficiency and falling easily into local optimal point in the optimization process. To improve the global optimization performance and local optimization efficiency of particle swarm optimization algorithm, this paper proposes a chaotic particle swarm optimization algorithm with adaptive mutation（ACPSO）by introducing a chaotic mapping and an adaptive strategy. Compared to five existing algorithms, numerical results on six benchmark test functions indicate that ACPSO algorithm has better performance.

关 键 词：粒子群优化 自适应策略 混沌映射 数值优化 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]