作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2012, Vol. 38 ›› Issue (04): 182-184. doi: 10.3969/j.issn.1000-3428.2012.04.059

• 人工智能及识别技术 • 上一篇    下一篇

并行协同差异进化算法研究

王 磊 a,林鸿飞 a,滕弘飞 a,b   

  1. (大连理工大学 a. 计算机科学与技术学院;b. 机械工程学院,辽宁 大连 116023)
  • 收稿日期:2011-07-08 出版日期:2012-02-20 发布日期:2012-02-20
  • 作者简介:王 磊(1987-),男,硕士研究生,主研方向:智能计算,布局优化;林鸿飞、滕弘飞,教授、博士生导师
  • 基金资助:

    国家自然科学基金资助项目(50575031, 50975033)

Research on Parallel Cooperative Coevolutionary Differential Evolution Algorithm

WANG Lei a, LIN Hong-fei a, TENG Hong-fei a,b   

  1. (a. School of Computer Science and Technology; 2. School of Mechanical Engineering, Dalian University of Technology, Dalian 116023, China)
  • Received:2011-07-08 Online:2012-02-20 Published:2012-02-20

摘要: 在协同差异进化(CCDE)算法和整体同步并行(BSP)计算模型的基础上,提出一种并行协同差异进化算法。采用Archive协同机制取代 CCDE原有的协同机制,有助于得到算法最优解,并使用BSP模型实现CCDE的并行计算。利用标准测试函数进行仿真实验,结果表明,该算法具有较高的计算效率和计算质量。

关键词: 并行计算, 协同差异进化, 大规模优化问题, 同步并行

Abstract: This paper presents a Parallel Cooperative Coevolutionary Differential Evolution(PCCDE) algorithm based on Cooperative Coevolutionary Differential Evolution(CCDE) and Bulk Synchronous Parallel(BSP) computing model. In this algorithm, Archive collaboration mechanism replaces the original mechanism in CCDE, the PCCDE uses the BSP model to implement the parallel computation. Simulation experimental results based on a set of widely used benchmark function show that the algorithm has superior calculation efficiency and quality.

Key words: parallel computation, Cooperative Coevolutionary Differential Evolution(CCDE), large-scale optimization problem, synchronous parallel

中图分类号: