An Intelligent Parameter Selection Method for Particle Swarm Optimization Algorithm
For the problem of particle swarm optimization parameters selection, a kind of intelligent method to optimum parameters selection using another particle swarm optimization algorithm is proposed. Firstly it analyzes the effect of each parameter on algorithm performance in detail. Then it takes parameter selection of PSO algorithm as a complex optimization problem, sets appropriate fitness function to describe optimization performance, and uses PSO-PARA algorithm to optimize the parameters selection method of PSO-OPT algorithm. Tests to the benchmark function show that these parameters are better than the experience parameters test results in the optimal fitness, the mean value of optimal fitness, convergence rate.
parameter optimization particle swarm optimization algorithm analysis of parameters
Yuntao Dai Liqiang Liu Ying Li
College of Science Harbin Engineering University Heilongjiang, China College of Automation Harbin Engineering University Heilongjiang, China
国际会议
昆明、丽江
英文
960-964
2011-04-15（万方平台首次上网日期，不代表论文的发表时间）