A Novel Method for Finding Good Local Guides in Multi-objective Particle Swarm Optimization

Abstract

In multi-objective particle swarm optimization (MOPSO) methods, selecting good local guides (the global best particle) for each particle of the population from a set of Pareto-optimal solutions has a great impact on the convergence and diversity of solutions. This paper introduces the particle angle division method as a new method for finding the global best particle for each particle of the population. The particle angle division method is implemented and is compared with adaptive grid method based on the same MOPSO for different test functions. The results show our strategy can improve convergence and diversity of MOPSO largely.