计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (18): 9-12.DOI: 10.3778/j.issn.1002-8331.2010.18.004

• 博士论坛 • 上一篇    下一篇

基于交叉变异策略的双种群差分进化算法

谭 跃1,2,谭冠政1,伍雪冬3   

  1. 1.中南大学 信息科学与工程学院,长沙 410083
    2.湖南城市学院 物理与电信工程系,湖南 益阳 413000
    3.江苏科技大学 电子与信息学院,江苏 镇江 212003
  • 收稿日期:2010-03-23 修回日期:2010-05-06 出版日期:2010-06-21 发布日期:2010-06-21
  • 通讯作者: 谭 跃

Dual population differential evolution algorithm based on crossover and mutation strategy

TAN Yue1,2,TAN Guan-zheng1,WU Xue-dong3   

  1. 1.School of Information Science and Engineering,Central South University,Changsha 410083,China
    2.Department of Physics and Telecom Engineering,Hunan City University,Yiyang,Hunan 413000,China
    3.School of Electronics and Information,Jiangsu University of Science and Technology,Zhenjiang,Jiangsu 212003,China
  • Received:2010-03-23 Revised:2010-05-06 Online:2010-06-21 Published:2010-06-21
  • Contact: TAN Yue

摘要: 为加强差分进化算法的全局搜索能力,提出了一种基于交叉变异策略的双种群差分进化算法(CMDPDE)。CMDPDE中,两个种群分别采用大小不同的缩放因子和交叉因子,在每代进化完毕后,对其中缩放因子和交叉因子较小的种群执行交叉或变异策略来寻找更优的个体,同时两个种群之间每10代进行一次信息交流。这种方式与单种群差分进化算法相比,可以通过双种群和交叉变异策略来增加解的多样性,使算法能在更大的范围内寻优。6个Benchmark函数的实验结果证明CMDPDE具有较好的寻优能力。

关键词: 交叉, 变异, 双种群, 差分进化

Abstract: Dual Population Differential Evolution algorithm based on Crossover and Mutation strategy(CMDPDE) is proposed to enhance global search ability of single population differential evolution.In CMDPDE,one population uses big scale factor and crossover factor,the other with small scale factor and crossover factor will execute crossover or mutation operations to search better individual after an evolution for each individual evolves one time per generation.At the same time evolution information will be exchanged between two populations after all individuals of two populations evolve ten times.Compared with single population differential evolution,CMDPDE increases diversity of solutions through dual population and crossover and mutation strategy,which makes CMDPDE search better solutions in a larger range.Experiment results on six benchmark functions show that CMDPDE has the better ability of finding optimal solution.

Key words: crossover, mutation, dual population, Differential Evolution(DE)

中图分类号: