Multi-objective random drift particle swarm optimization algorithm with adaptive grids

Abstract

In this paper, we propose a multi-objective random drift particle swarm optimization algorithm with adaptive grids (MORDPSO-AG) to solve the multi-objective optimization problem. Due to the good search performance of the RDPSO, the proposed algorithm can find more accurate Pareto optimal solutions quickly. However, like PSO and other population-based search techniques, the loss of diversity and premature convergence are inevitable. Therefore, we introduce the method of adaptive grids into RDPSO to maintain the swarm diversity. We adopt an external archive to reserve the found Pareto optimal solutions, and update the solutions based on adaptive grids. Besides, in order to make the lead particle guide the particle swarm to find the true Pareto optimal solutions, we select the leader particle by using roulette wheel method. Fianlly, we use four benchmark test functions to evaluate the performance of the algorithm, and the experimental results show that the proposed algorithm has better convergence and solution distribution than the other tested methods.