计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (30): 53-57.

• 研究、探讨 • 上一篇    下一篇

带极值抖动的变尺度粒子群优化算法

刘 进1,覃洁萍2   

  1. 1.广西师范学院,南宁 530023
    2.广西中医学院,南宁 530001
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-10-21 发布日期:2011-10-21

Mutative scale particle swarm optimization algorithm with extremum disturbed

LIU Jin1,QIN Jieping2   

  1. 1.Guangxi Teachers Education University,Nanning 530023,China
    2.Guangxi Traditional Chinese Medicine University,Nanning 530001,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-10-21 Published:2011-10-21

摘要: 为克服粒子群优化算法容易陷入局部最优解的问题,提出一种带极值抖动的变尺度粒子群优化算法,该算法在粒子进化过程中动态调整学习因子,改善粒子的搜索性能,利用极值抖动方法帮助粒子逃离局部最优解,采用变尺度方法逐步缩小算法的优化范围,提高算法搜索密度。实验表明,该算法对9个具有代表性的基准测试函数,其优化效率及优化精度均优于以往提出的典型粒子群优化改进算法。

关键词: 粒子群优化算法, 极值抖动, 变尺度

Abstract: To overcome the drawbacks of usually being trapped into local minima when solving complex optimal problem using the algorithm of particle swarm optimization,a mutative scale particle swarm optimization algorithm with extremum disturbed is proposed.In the procedure of the particles evolution,the proposed algorithm dynamically adjusts learning factors for improving the search performance,introduces in extremum disturbed method to help the particles escape unexpected local minima,adopts the method of mutative scale shortening the optimal range to increase search density.Experiment results show that the proposed algorithm is superior to previously improved PSO algorithm in efficiency and precision on 9 well-known benchmark test functions.

Key words: Particle Swarm Optimization(PSO), extremum disturbed, mutative scale