Chaos Adaptive Improved Particle Swarm Algorithm for Solving Multi-Objective Optimization

Abstract

To overcome the problem of premature convergence on Particle Swarm Optimization (PSO), this paper proposes both the improved particle swarm optimization methods (IPSO) based on self-adaptive regulation strategy and the Chaos Theory. Given the effective balance of particles’ searching and development ability, the self-adaptive regulation strategy is employed to optimize the inertia weight. To improve efficiency and quality of searching, the learning factor is optimized by generating Chaotic Sequences by Chaos Theory. The improved method proposed in this paper achieves better convergence performance and increases the searching speed. Simulation results of some typical optimization problems and comparisons with typical multi-objective optimization algorithms show that IPSO has a fast convergence speed, the diversity of non-dominated and the ideal convergence. The algorithm meets the requirements of multi-objective optimization problem.

DOI : http://dx.doi.org/10.11591/telkomnika.v12i1.3189