A New Approach to Improve Particle Swarm Optimization

Abstract. Particle swarm optimization (PSO) is a new evolutionary computation technique. Although PSO algorithm possesses many attractive properties, the methods of selecting inertia weight need to be further investigated. Under this consideration, the inertia weight employing random number uniformly distributed in [0,1] was introduced to improve the performance of PSO algorithm in this work. Three benchmark functions were used to test the new method. The results were presented to show that the new method is effective.

• Type: Conference
• Year: 2003
• Where: GECCO
• Authors: Liping Zhang, Huanjun Yu, Shangxu Hu
• Keywords: GECCO 2003, Inertia Weight, Particle Swarm Optimization, PSO Algorithm