计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (6): 60-62.DOI: 10.3778/j.issn.1002-8331.2009.06.018

• 研究、探讨 • 上一篇    下一篇

基于动态邻域的粒子群算法的研究

韩立娜,熊盛武   

  1. 武汉理工大学 计算机学院,武汉 430070
  • 收稿日期:2008-01-15 修回日期:2008-04-02 出版日期:2009-02-21 发布日期:2009-02-21
  • 通讯作者: 韩立娜

Research on particle swarm optimization based on dynamic neighborhood

HAN Li-na,XIONG Sheng-wu   

  1. School of Computer Science & Technology,Wuhan University of Technology,Wuhan 430070,China
  • Received:2008-01-15 Revised:2008-04-02 Online:2009-02-21 Published:2009-02-21
  • Contact: HAN Li-na

摘要: 研究了邻域拓扑结构对粒子群算法性能的影响。设计了两种动态邻域生成策略,并基于一组具有代表性的测试函数,对两种典型的算法模型——标准的粒子群算法(CPSO)和充分联系的粒子群算法(FIPS)进行实验。实验结果表明,不同的邻域拓扑结构和不同的算法模型都能够影响粒子群算法的性能。

关键词: 粒子群, 动态邻域, 生成策略

Abstract: The neighborhood topology’s influence on the performance of a PSO is analyzed.Two methods to generate dynamic neighborhood are proposed.Based on a representative set of benchmark functions,the proposed scheme is tested.The result of numerical experiments demonstrate that the performance of a particle swarm depends on population topology as well as algorithm version.

Key words: particle swarm optimization, dynamic neighborhood, generating method