Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (22): 44-46.

• 学术探讨 • Previous Articles     Next Articles

Multi-particle swarm co-evolution algorithm

LI Fei-fei1,YAO Kun1,LIU Xi-yu2   

  1. 1.School of Information Science & Engineering,Shandong Normal University,Ji’nan 250014,China
    2.School of Management,Shandong Normal University,Ji’nan 250014,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-08-01 Published:2007-08-01
  • Contact: LI Fei-fei

一种多微粒群协同进化算法

李菲菲1,姚 坤1,刘希玉2   

  1. 1.山东师范大学 信息科学与工程学院,济南 250014
    2.山东师范大学 管理学院,济南 250014
  • 通讯作者: 李菲菲

Abstract: Illumined by phenomenon of co-evolution in nature,particle swarm optimization is combined with co-evolution,a multi-particle swarm co-evolution algorithm is presented.In the process of evolution,particle not only exchanges information with other particles in its sub-swarm but also is influenced by other sub-swarms.By doing experiments on three benchmark functions,the results show that the algorithm avoids to trap into local optimum in a certain extent and improves the precision of convergence.

Key words: particle swarm optimization, co-evolution, multi-particle swarm co-evolution

摘要: 受自然界共生现象的启发,将微粒群算法和协同进化相结合,提出了一种多微粒群协同进化算法。进化过程中,粒子不仅要与本子群的其他微粒交换信息,还要受其他子群体的影响。通过对三个标准函数优化的实验结果表明,此算法在一定程度上避免了陷入局部极值点并且提高了收敛精度。

关键词: 微粒群算法, 协同进化, 多微粒群协同进化算法