计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (10): 38-40.DOI: 10.3778/j.issn.1002-8331.2010.10.013

• 研究、探讨 • 上一篇    下一篇

设立禁区的多粒子群优化算法

高 翔,唐普英   

  1. 电子科技大学 光电信息学院,成都 610054
  • 收稿日期:2008-10-08 修回日期:2009-01-04 出版日期:2010-04-01 发布日期:2010-04-01
  • 通讯作者: 高 翔

Algorithm of multi-PSO with forbidden zone

GAO Xiang,TANG Pu-ying   

  1. School of Optoelectronic Information,University of Electronic Science and Technology of China,Chengdu 610054,China
  • Received:2008-10-08 Revised:2009-01-04 Online:2010-04-01 Published:2010-04-01
  • Contact: GAO Xiang

摘要: 提出一种设立禁区的多粒子群优化(FZPSO)方法。FZPSO模仿某些鸟群的搜索方法,即用多数的鸟群搜索大部分空间,而用少数的鸟群搜索特定小空间,用来追逐当前全局最优解以加快算法收敛。以禁区的概念来改善粒子群容易陷入局部最优的弱点。禁区的大小由禁区半径决定,将每个局部最优解为中心,半径以内的区域划为禁区。实验结果表明FZPSO具有较强的全局收敛能力。

关键词: 粒子群算法, 函数优化, 禁区, 禁区半径

Abstract: A method of multi-PSO with Forbidden Zone(FZPSO) is presented.FZPSO imitates some birds search method,which uses most of the birds search wide range space,and uses a small number of birds search for specific small space to chase the current global best solution to speed up the convergence.By using the concept of the forbidden zone,it is able to change the weakness of the PSO easily fall in the local best solution.Forbidden zone is decided by the size of the forbidden radius,each local optimal solution as the center and the radius of the area designated as forbidden zones.The results show that FZPSO has a strong global convergence capability.

Key words: particle swarm optimizer, function optimization, forbidden zone, forbidden radius

中图分类号: