计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (36): 16-19.DOI: 10.3778/j.issn.1002-8331.2008.36.004

• 博士论坛 • 上一篇    下一篇

基于信息熵的异类多种群蚁群算法

邓 可1,林 杰1,张 鹏1,2   

  1. 1.同济大学 经济与管理学院,上海 200092
    2.西安理工大学 信息系,西安 710048
  • 收稿日期:2008-07-31 修回日期:2008-10-13 出版日期:2008-12-21 发布日期:2008-12-21
  • 通讯作者: 邓 可

Multiple heterogeneous ant colonies algorithm based on information entropy

DENG Ke1,LIN Jie1,ZHANG Peng1,2   

  1. 1.Department of Economics & Management,Tongji University,Shanghai 200092,China
    2.Department of Information Engineering,Xi’an University,Xi’an 710048,China
  • Received:2008-07-31 Revised:2008-10-13 Online:2008-12-21 Published:2008-12-21
  • Contact: DENG Ke

摘要: 提出了一种基于信息熵的异类多种群蚁群算法。算法使用多个异类种群的蚂蚁子群体同时进行优化计算,引入信息熵来表示蚂蚁种群的进化程度,根据蚂蚁子群体间的信息熵来决定子群体间的信息交流策略,包括选择信息交流的对象和调节信息交流的周期以及信息更新策略,以取得各蚂蚁子群体中解的多样性和收敛性之间的动态平衡。基于旅行商问题的实验证明,该算法具有很好的全局搜索能力、收敛速度以及解的多样性。

关键词: 信息熵, 异类多种群, 蚁群算法, 旅行商问题

Abstract: This paper presents an Information Entropy Based Multiple Heterogeneous Ant Colonies Algorithm(IEBMHACA).This algorithm uses multiple heterogeneous ant colonies in simultaneous optimization calculation.For each ant colony,evolution degree is expressed by information entropy,so the ants can adaptively choose strategies to exchange information between colonies to achieve dynamic equilibrium between solution diversity and convergence speed.Those strategies include selecting information exchanging object,regulating exchanging cycle and updating pheromone.Experimental results on traveling salesman problem show that this algorithm has a good global searching ability,high convergence speed and solution diversity.

Key words: information entropy, multiple heterogeneous, ant colonies algorithm, Travel Salesman Problem(TSP)