基于搜索空间可调的自适应粒子群优化算法与仿真

Adaptive particle swarm optimization based on search space adjustable and simulation

针对收缩因子粒子群优化(CPSO)算法易陷入局部最优和发生过早收敛的问题,提出了基于搜索空间可调的自适应粒子群优化(APSO)算法.该算法根据种群早熟收敛程度和个体适应值,在CPSO算法停滞时,将全部粒子有效地划分在3类不同的搜索空间,使种群始终保持搜索空间的多样性,易于跳出局部最优,从而有效地改善了CPSO算法后期的寻优能力.

An adaptive particle swarm optimization （APSO） algorithm based on the search space adjustable is proposed and applied to solve the problem that constriction-factor PSO （CPSO） algorithm easily fall into local optimal and occur premature convergence. When the CPSO algorithm stagnates, according to the extent premature convergence groups and individual fitness, the algorithm will divide particles to three different searching spaces, by which the swarm is kept to maintain the diversity of the searching space and easy to jump out of local optima. The late CPSO algorithm optimization capabilities availability are improved.