计算机与现代化 ›› 2009, Vol. 1 ›› Issue (10): 18-3.doi:

• 论文 • 上一篇    下一篇

蚁群算法的并行性研究

苏盛曲,梁声灼   

  1. 南昌大学信息工程学院,江西 南昌 330031
  • 收稿日期:2008-09-28 修回日期:1900-01-01 出版日期:2009-10-15 发布日期:2009-10-15

Parallel Study of Ant Colony Algorithm

SU Sheng-qu,LIANG Sheng-zhuo   

  1. School of Information Engineering, Nanchang University, Nanchang 330031, China
  • Received:2008-09-28 Revised:1900-01-01 Online:2009-10-15 Published:2009-10-15

摘要: 并行计算作为现代计算机的一种重要的计算方法,在很大程度上优化了蚁群算法的计算过程。蚁群算法本身隐含着一定的并行性,从本质上来说,蚁群算法是以并行式的协同优化计算方式为特征,利用并行计算求出最优解。本文重点讨论蚁群算法的并行实现,并通过一个仿真实验验证并行优化蚁群算法在解决一个具有时变动态、连续、多输入、非线性系统的最优控制问题上的最优解决方法,得出蚁群算法在加速比上更具有优势。

关键词: 蚁群算法, 并行计算, 最优控制

Abstract: Parallel computing as modern computer an important method of calculation, in a large part, optimizes the ant colony algorithm calculation processes. The ant colony algorithm itself implies a parallelism, in essence, the ant algorithm is characterized by the parallel collaborative optimization calculation methods and gets optimization solution by it. This article focuses on realization the ant colony algorithm parallel, through a simulation to verify that the ant colony optimization algorithm in parallel is the optimal solution method in solving problems with timevarying dynamic, continuous, multiinput and the optimal control of nonlinear system then arrive a conclusion that the ant algorithm has more advantages than anther in accelerator.

Key words: ant colony algorithm, parallel computing, optimal control