Hybrid Particle Swarm Optimisation Algorithms Based on Differential Evolution and Local Search

摘要

Particle Swarm Optimisation (PSO) is an intelligent search method based on swarm intelligence and has been widely used in many fields. However it is also easily trapped in local optima. In this paper, we propose two hybrid PSO algorithms: one uses a Differential Evolution (DE) operator to replace the standard PSO method for updating a particle's position; and the other integrates both the DE operator and a simple local search. Seven benchmark multi-modal, high-dimensional functions are used to test the performance of the proposed methods. The results demonstrate that both algorithms perform well in quickly finding global solutions which other hybrid PSO algorithms are unable to find.