基于个体最优位置的自适应变异扰动粒子群算法

针对粒子群算法在寻优时容易陷入局部最优的不足,提出了一种基于个体最优位置的自适应变异扰动粒子群算法AMDPSO (adaptive mutation disturbance particle swarm optimization).该算法以粒子群算法为基础,加入扰动,当满足自适应条件时,粒子以个体最优位置为依据进行变异操作.将该算法运用于6个测试函数,并与惯性权重粒子群算法、收缩因子粒子群算法以及差分进化算法进行了比较,结果表明:AMDPSO能在寻优过程中让粒子跳出局部最优,保持种群多样性,具有更好的收敛速度和优化性能. 

Adaptive Mutation Disturbance Particle Swarm Optimization Algorithm Based on Personal Best Position

In order to overcome the disadvantage of the particle swarm optimization(PSO) that it easily falls into local optimum,an adaptive mutation disturbance particle swarm optimization(AMDPSO) algorithm based on personal best position was proposed.This algorithm is based on PSO,and the disturbance is considered.When the adaptive conditions are met,the mutation operation of particles is performed based on the personal best position.The proposed algorithm was applied to 6 test functions and compared with IWPSO(inertia weight particle swarm optimization),CFPSO(constriction factor particle swarm optimization) and DE(differential evolution).The research results show that the AMDPSO has a good convergence rate and optimization capability,and can easily escape the local optimum and keep the population diversity.