Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (12): 66-68.

• 理论研究 • Previous Articles     Next Articles

Differential evolution algorithm based on chaos searching

LIU Jun-min1,GAO Yue-lin2   

  1. 1.School of Mathematics and Computer,Ningxia University,Yinchuan 750021,China
    2.Research Institute of Information and System Computation Science,North National University,Yinchuan 750021,China
  • Received:2007-08-13 Revised:2007-11-14 Online:2008-04-21 Published:2008-04-21
  • Contact: LIU Jun-min

基于混沌搜索的微分进化算法

刘军民1,高岳林2   

  1. 1.宁夏大学 数学与计算机学院,银川 750021
    2.北方民族大学 信息与系统科学研究所,银川 750021
  • 通讯作者: 刘军民

Abstract: A Chaos Differential Evolution algorithm(CDE) is introduced to overcome the problem of low convergence speed and bad searching ability in the later evolution period.CDE makes good use of the ergodicity,stochastic property,regularity of chaos.The proposed algorithm has not only kept the simple of original differential evolution algorithm,but also improved the rapidity of convergence,the precision of computational and the ability of global optimization.The experiment results demonstrate that the new algorithm proposed is superior to the original differential evolution algorithm.

Key words: global optimization, differential evolution algorithm, chaos

摘要: 针对基本微分进化算法在后期收敛速度慢,搜索能力差等问题,利用混沌搜索的随机性、遍历性以及对初值的敏感性等特性,提出了一种混合混沌搜索的微分进化算法——混沌微分进化算法。该算法既保持了基本微分进化算法结构简单的特点,又能提高算法的收敛速度、计算精度以及全局寻优能力。数值仿真结果表明,该算法的性能优于基本微分进化算法。

关键词: 全局优化, 微分进化算法, 混沌