A Modified Particle Swarm Optimization Algorithm with Design of Experiment Technique and a Perturbation Process

摘要

Particle swarm optimization (PSO) is a relatively new stochastic optimization algorithm and has gained much attention in recent years because of its fast convergence speed and strong optimization ability. However, PSO suffers from premature convergence problem for quick losing of diversity. That is to say, if no particle discovers a new superiority position than its previous best location, PSO algorithm will fall into stagnation and output local optimum result. In order to improve the diversity of basic PSO, design of experiment technique is used to initialize the particle swarm in consideration of its space-filling property which guarantees covering the design space comprehensively. And the optimization procedure of PSO is divided into two stages, optimization stage and improving stage. In the optimization stage, the basic PSO initialized by Optimal Latin hypercube technique is conducted. Based on the result of the optimization stage, a perturbation course is used to release the particles out from stagnation in the improving stage. According to these methods, a modified PSO algorithm, namely OLPPSO (Optimal Latin Hypercube design and a perturbation process are used to enhance basic PSO) is proposed. The proposed method is tested and validated by standard benchmark functions in contrast with the basic PSO. Based on the experimental results, the OLPPSO algorithm outperforms the basic PSO by noticeable percentage.