• 论文 •    

基于MapReduce的蚁群算法

吴昊,倪志伟,王会颖,   

  1. 1.合肥工业大学 管理学院,安徽合肥230009;2.合肥工业大学 教育部过程优化与智能决策重点实验室,安徽合肥230009;3.安徽财贸职业学院 电子信息系,安徽合肥230601
  • 出版日期:2012-07-25 发布日期:2012-07-25

MapReduce-based ant colony optimization

WU Hao, NI Zhi-wei, WANG Hui-ying,   

  1. 1.School of Management, Hefei University of Technology, Hefei 230009, China; 2.Key Laboratory of Process Optimization and Intelligent Decision-making, Ministry of Education,Hefei University of Technology, Hefei 230009, China;3.Department of Electronics and Information, Anhui Finance & Trade Vocational College, Hefei 230601,China
  • Online:2012-07-25 Published:2012-07-25

摘要: 云计算环境下应用蚁群算法分布式并行对问题进行求解的研究较少,且蚁群算法存在搜索时间长和易收敛于非最优解的缺陷,当问题的规模较大时求解困难。为此应用云计算技术将蚁群算法并行化,提出基于MapReduce的蚁群算法。该算法将分治思想和模拟退火算法融入蚁群算法,改进其缺陷,并应用于求解较大规模的旅行商问题。仿真实验取得了较好的效果,且获得了测试实例gr666的新解。

关键词: 云制造, 云计算, 蚁群算法, 分治, 模拟退火算法

Abstract: The researches on solving problems with Ant Colony Optimization(ACO)distributed parallel under cloud computing were less, and ACO had defects in long search time and convergence in non-optimal solution. When the scale of problem was large, it was too hard to solve. Therefore, MapReduce-based ACO was proposed by using cloud computing to parallel ACO. In this algorithm, dividing conquer and simulated annealing algorithm were merged into ACO to improve the defects. It was also applied to solve large-scale of Traveling Salesman Problem(TSP). The simulation experiment got a well effect and the new solutions of test gr666 were obtained.

Key words: cloud manufacturing, cloud computing, ant colony optimization, divide conquer, simulated annealing algorithm, traveling salesman problem

中图分类号: