Empirical Study of a Particle Swarm Optimization with Extended Memory

摘要：

Particle swarm optimizer (PSO) is susceptible to poor convergence and weak robustness especially at high dimensions. To address this problem, a particle swarm optimizer with extended memory (PSOEM) is proposed in this paper. The extended memory is imported to store each particle historical information including recent personal (local) best positions and global best positions. Moreover, this algorithm employs a parameter to describe the importance of extended memory in different stages of evolution process. The specialty of the presented algorithm is that it can integrate with numerous existing improved PSO algorithms. With some well-known benchmark functions, a numeric comparison between the PSO and the PSOEM in terms of convergence speed and robustness is undertaken at corresponding points. The experimental results show that the convergence speed of the proposed algorithm is superior to the basic PSO. Besides, robustness is improved remarkably.

关键词： Particle Swarm Optimizer Extended Memory PSOEM

作者: Lei Lei Zhou Lai-yuan Zhang Cong-li Huang Da-wei

作者单位: College of AutomationChongqing UniversityChongqing, China College of Automation Chongqing University Chongqing, China

会议类型: 国际会议

会议名称: 2011 International Conference on Information System and Computational Intelligence(2011 IEEE信息系统与计算智能国际会议 ICISCI 2011)

会议地点: 哈尔滨

会议语种:英文

页码: 440-444

在线出版日期: 2011-01-18（万方平台首次上网日期，不代表论文的发表时间）