三群协同粒子群优化算法

Three Swarms Cooperative Particle Swarm Optimization

针对基本粒子群优化算法易陷入局部极值点、搜索精度低等缺点,提出了一种三群协同粒子群优化算法(TSC-PSO)。搜索时,如果全局极值连续若干代没有改善,粒子未找到全局最优点,就任选某个优群,将其群内粒子和差群粒子交换。仿真结果显示,对一些经典多峰值函数、非凸病态函数,TSC-PSO增强了全局搜索能力,具有比基本PSO更好的优化性能。

In order to overcome the drawback of basic PSO,such as being subject to falling into local optimization and being poor in performance of precision, an improved PSO algorithm, three swarms cooperative particle swarm optimization （TSC-PSO）, is proposed. Regarding to several special multimodal functions and singular non-convex functions, the results of simulation show that the TSC-PSO can strengthen the global searching ability and have better optimization performance than basic PSO.