摘要
粒子群优化算法PSO目前仍存在着早熟收敛和收敛速度较慢的难题,提出一种新的PSO改进算法。该算法利用水平集对PSO的每一代粒子按照适应度划分成两个子种群,对两种群采用不同进化策略,并适当进行信息交换,通过采用这种策略提高了算法的收敛速度和精度,同时也减少早熟发生的机会。实验证明,这种改进的算法是非常有效的。
Recently there still exist some problems in particle swarm optimization （PSO） algorithm including prematurity and slow convergence.To solve these problems, an improved PSO based on level set is presented. This algorithm used level set to divided particles of each generation into two child populations according to their fitness, then the two child populations use different evolutionary strategy to evolve and exchange information appropriately, so this algorithm accelerate the convergence speed , enhanced accuracy of the algorithm and reduce the chance of premature. Experiment shows that this algorithm is very effective.
出处
《微计算机信息》
2011年第12期144-146,96,共4页
Control & Automation
基金
河南理工大学青年基金项目(Q2011-32)
关键词
粒子群优化算法
动态
维信息
水平集
多种群
particle swarm optimization algorithm
dynamic
dimension information
level set
multi-specie
