Improved Multi-Objective PSO algorithm for Optimization Problems

Abstract

Some Particle Swarm Optimization (PSO) algorithm have been used to solve Multi-Objective Optimization Problems (MOP) and have achieved good results. But finding a good convergence and distribution of solutions near the Pareto-optimal front in little computational time is still a hard work especially for some complex functions. This paper introduces an improved multi-objective PSO algorithm. It is called Strength Pareto Particle Swarm Optimization algorithm(SPPSO) which uses the ranking and sharing strategies of Strength Pareto Evolutionary Algorithm II (SPEA2). The hyper-volume metric (Zitzler 1999) is introduced to evaluate overall performance of the obtained solutions. Simulation results on five difficult test problems show that the proposed algorithm is able to find much better spread of solutions and better convergence near the true Pareto-optimal front compared to CMOPSO.