计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (26): 28-30.DOI: 10.3778/j.issn.1002-8331.2010.26.009

• 研究、探讨 • 上一篇    下一篇

改进分布估计算法的协同优化研究

吴 红,王维平,王 磊,杨 峰   

  1. 国防科学技术大学 信息系统与管理学院,长沙 410073
  • 收稿日期:2009-07-20 修回日期:2009-12-01 出版日期:2010-09-11 发布日期:2010-09-11
  • 通讯作者: 吴 红

Research on cooperative optimization of improved estimation of distribution algorithm

WU Hong,WANG Wei-ping,WANG Lei,YANG Feng   

  1. School of Information System and Management,National University of Defense Technology,Changsha 410073,China
  • Received:2009-07-20 Revised:2009-12-01 Online:2010-09-11 Published:2010-09-11
  • Contact: WU Hong

摘要: 针对分布估计算法在问题求解过程中容易陷入局部最优的缺点,引入物理退火的思想,提出模拟退火的分布估计算法,其中退火温度以信息熵表示。然后以此为基础,面向多核处理器提供的并行计算能力,提出多量子分布估计的协同优化算法。仿真实验表明,该算法缩短了优化时间,提高了优化结果。

关键词: 分布估计算法, 物理退火, 信息熵, 协同优化

Abstract: For that EDA is easy to fall in local optimum in the process of problem solving,the physical annealing is introduced into this algorithm and the Simulated Annealing Estimation of Distribution algorithm(SAEDA) is put forward,annealing temperature denoted by information entropy.Afterward,based on this,for parallel compute ability proved by multi-core processor, a quantum-behave estimation of distributions cooperative optimization algorithm is put forward.The simulation experiment result shows that the algorithm can shorten optimize time and improve optimize result.

Key words: estimation of distribution algorithm, physical annealing, information entropy, cooperative optimization

中图分类号: