基于MPI的并行蚁群算法的研究与实现

广东工业大学学报 ›› 2008, Vol. 25 ›› Issue (1): 38-42.

基于MPI的并行蚁群算法的研究与实现

广东外语外贸大学南国商学院；华南理工大学计算机学院；华南理工大学计算机学院广东广州510545；广东广州510640；广东广州510545；广东工业大学经济管理学院；广东广州510520；

出版日期:2008-01-01 发布日期:2008-01-01
基金资助:
广东省自然科学基金资助项目(05011896);广东省教育厅自然科学基金资助项目(Z03080)

Research and Implementation of Parallel Ant Colony Optimization Algorithm Based on MPI

(1.Nanguo Business College,Guangdong University of Foreign Studies,Guangzhou 510545,China;2.School of Computer Science,South China University of Technology,Guangzhou 510640,China;3.School of Economics and Management,Guangdong University of Technology,Guangzhou 510520,China)

Online:2008-01-01 Published:2008-01-01

摘要/Abstract

摘要： 在消息传递接口(message passing interface,MPI)的基础上,采用划分蚁群的策略,实现了基于MPI的并行蚁群算法,并对该算法采用旅行商问题进行了实验.实验结果表明,使用并行计算技术,可以很好地提高运行速度.

关键词: 蚁群算法；消息传递接口；旅行商问题；

Abstract: In this paper,parallel strategy is used develop parallel ant colony optimization algorithm based on Message Passing Interface.The Traveling Salesman Problem is experimented using our system.The results demonstrate that our algorithm is superior to the existing ant colony optimization algorithm and has better running time by using parallel computing.

Key words: ant colony optimization algorithm； message passing interface(MPI)； traveling salesman problem(TSP)；

刘东；常静；魏文红；赵洁； . 基于MPI的并行蚁群算法的研究与实现[J]. 广东工业大学学报, 2008, 25(1): 38-42.

Liu Dong1,2,Chang Jing1,2,Wei Wen-hong2,Zhao Jie2,3. Research and Implementation of Parallel Ant Colony Optimization Algorithm Based on MPI[J]. Journal of Guangdong University of Technology, 2008, 25(1): 38-42.

参考文献

[1] 张静乐,王世卿,王乐. 具有新型遗传特征的蚁群算法[J]. 微计算机信息. 2006(05)

[1] Dorigo M,,Maniezzo V,Colorni A.The ant system:optimiza-tion by a colony of cooperating agents. IEEE Transca-tions on Systems,Man and Cybernetics . 1996

[2] Chu S C,Roddick J F,Pan J S,et al.Parallel ant colonysystems. Lecture Notes in Artificial Intelligence . 2003

[3] Dorigo M,Gambardella L M.A Cooperative Learning Approach to the Traveling Salesman Problem. IEEE Transactions on Evolutionary Computation . 1997

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed