An Improved PSO Algorithm Based on Mutation Operator and Simulated Annealing

Abstract

Particle swarm optimization (PSO) algorithm is simple stochastic global optimization technique, but it exists unbalanced global and local search ability, slow convergence speed and solving accuracy. An improved simulated annealing (ISAM) algorithm is introduced into the PSO algorithm with crossover and Gauss mutation to propose an improved PSO (ISAMPSO) algorithm based on the mutation operator and simulated annealing in this paper. In the ISAMPSO algorithm, the mutation operator of genetic algorithm is introduced into the SA algorithm as a generation mechanism of new solution in order to propose an improved simulated annealing algorithm with mutation (ISAM). Then the ISAM algorithm is introduced into the PSO algorithm to jump out the local optimum, effectively achieve the global optimum adjust and optimize the population, maintain the diversity of the population, improve the local search ability and convergence speed. Six classical functions are selected to test the performance of the proposed ISAMPSO algorithm. The simulation experiments results show that the proposed ISAMPSO algorithm can effectively overcomes the stagnation phenomenon and enhance the global search ability. The convergence speed and accuracy were better than the PSO algorithm.