Analysis Of Parameters For Multi-swarm Cooperative Particle Swarm Optimizer
Motivated by the phenomenon of symbiosis in natural ecosystems, multi-swarm cooperative particle swarm optimization (MCPSO) is proposed. The parameters selection of MCPSO is a key factor to determine the performance of MCPSO. Traditionally, the parameters settings are often based on users experience, resulting in work load and hard to get the optimal parameters, which affected the usability of the algorithm. This paper presents a fixed set of key parameters of MCPSO. It is amply demonstrated by applying it for four benchmark functions. The experimental results demonstrate that our parameters selection strategy can obviously improved MCPSO algorithm performance, convergence rate and solution precision.
MCPSO optimization particle swarm
NIU Ben RAO Junjun TAN Lijing
Hefei Institute of Intelligent Machines. Chinese Academy of Sciences, Hefei. China, 230031 College o Hefei Institute of Intelligent Machines. Chinese Academy of Sciences, Hefei. China, 230031 Measurement Specialties Inc, Shenzhen, China, 518060
国际会议
2010 International Conference on Management(2010管理国际大会)
上海
英文
81-87
2010-07-24（万方平台首次上网日期，不代表论文的发表时间）