一种具有随机变异性质的粒子群协同优化算法

Improved Particle Swarm Cooperative Optimization Algorithm with Stochastic Mutation

给出了一种具有随机变异特性的改进型粒子群协同优化算法,该算法克服了传统粒子群算法易陷入局部最优解且后续迭代过程速度慢的缺点.在迭代过程中,粒子的变异概率取决于粒子的适应度值以及当前所有粒子的聚集度.通过变异,粒子可有效地探索新的空间领域,从而可以有效地避免陷入局部最优解.Benchmark函数实验结果表明,优化后的粒子群算法比传统粒子群算法具有较快的收敛速度和较高的全局收敛能力.

This paper proposes an improved particle swarm cooperative optimizer with stochastic mutation （SMPSCO） to solve the problems of easily falling into local optimum solution and slow convergence speed of the traditional particle swarm optimization （PSO）. During the iterating process, the mutation probability of the current particle depends on the fitness value and the gathering degree of particles. The exploration ability is efficiently improved by the mutation, and the probability of falling into local optimum is greatly decreased. The experimental results of the benchmark functions show the SMPSCO has faster convergence speed and higher global convergence ability than the traditional PSO.