Computer Engineering and Applications ›› 2012, Vol. 48 ›› Issue (14): 33-37.

Previous Articles     Next Articles

Parallel glowworm swarm optimization algorithm with master-slave structure

LIU Jiakun, ZHOU Yongquan   

  1. School of Mathematics and Computer Science, Guangxi University for Nationalities, Nanning 530006, China
  • Online:2012-05-11 Published:2012-05-14

具有主从结构的并行人工萤火虫群优化算法

刘佳昆,周永权   

  1. 广西民族大学 数学与计算机科学学院,南宁 530006

Abstract: Since glowworm swarm optimization algorithm is easy to fall into local optimization and has the low speed of convergence, it puts forward a parallel glowworm swarm optimization with master-slave structure by dividing the processing object into several sub-groups. Through eight typical functions testing, it is shown that the proposed algorithm has better performance in terms of decreasing computing time and avoiding falling into local optimization, thus obviously improving the optimization global ability.

Key words: Glowworm Swarm Optimization(GSO), master-slave structure, parallel algorithm, functions optimization

摘要: 基本人工萤火虫算法存在着易陷入局部极小、收敛速度慢等缺点。将算法的处理对象分为若干个子群,采用“雇主/工人”结构,提出了一种具有主从结构的并行人工萤火虫群优化算法。通过8个典型函数测试,实验结果表明所提出的算法在减少计算时间和避免陷入局部最优等方面具有较好的表现,从而提高了人工萤火虫算法整体性能。

关键词: 人工萤火虫算法, 主从结构, 并行算法, 函数优化