计算机工程与应用 ›› 2009, Vol. 45 ›› Issue (32): 41-44.DOI: 10.3778/j.issn.1002-8331.2009.32.013

• 研究、探讨 • 上一篇    下一篇

改进差分进化策略在多峰值函数优化中的应用

夏慧明1,周永权2   

  1. 1.南京师范大学 泰州学院 数学系,江苏 泰州 225300
    2.广西民族大学 数学与计算机科学学院,南宁 530006
  • 收稿日期:2008-12-12 修回日期:2009-02-20 出版日期:2009-11-11 发布日期:2009-11-11
  • 通讯作者: 夏慧明

Improved differential evolution strategy optimization algorithm for multiple hump functions

XIA Hui-ming1,ZHOU Yong-quan2   

  1. 1.Department of Mathematics,Taizhou College,Nanjing Normal University,Taizhou,Jiangsu 225300,China
    2.College of Mathematics and Computer Science,Guangxi University for Nationalities,Nanning 530006,China
  • Received:2008-12-12 Revised:2009-02-20 Online:2009-11-11 Published:2009-11-11
  • Contact: XIA Hui-ming

摘要: 针对差分演化算法与进化策略算法中所存在的不足,将模拟退火算子引入到差分演化算法的变异操作中,这样有助于在进化前期进行全局搜索,后期进行局部搜索;在标准进化策略的基础上,加入差分变异操作,提出了一种新的差分进化策略双重变异算法。通过测试算例可看出,该方法在多峰值函数优化问题中,具有求解精度较高,收敛速度较快等特点。

关键词: 退火因子, 差分演化算法, 进化策略, 双重变异

Abstract: Against to the finite about differential evolution algorithm and evolution strategy,this paper brings the simulated evolutionary operator into the differential evolution algorithm,it can help enhance global search in prophase and partial search at later period when evolving.Based on the normal evolution strategy adding the differential mutation operator in it,a new Bi-mutation differential evolution strategy algorithm is proposed.From the following examples,it can be seen that the result of the multiple hump function is very accurate and the convergence speed is fast.

Key words: anneal operator, differential evolution algorithm, evolution strategy, Bi-mutation

中图分类号: