An Improved Two-Swarm Based Particle Swarm Optimization Algorithm

Abstract

Basic particle swarm optimization (PSO) algorithm are susceptible to being trapped into local optimum and premature convergence happens. Inspired by the idea of genetic algorithm (GA), a new two-swarm based PSO algorithm (TSPSO) with roulette wheel selection is proposed. With different parameter settings, the two swarms have different flying trajectory, explore solution space as possible as they can, and enhance the global exploration ability. Roulette-wheel-selection based stochastic selection scheme make particles searching in the neighborhood of better feasible solution intensively and enhances the local exploitation ability. The proposed algorithm is tested on three benchmark test functions. The results show that the proposed algorithm is superior to PSO and GA in the solution of complex optimization problems