A Novel Scheme for Particle Swarm Optimization
Wei He, Yuanming Xu, Yaoming Zhou, Zhijun Meng, Yuankai Li · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2015
Particle Swarm Optimization (PSO) is one of the most popular optimization algorithms, but it suffers from the contradiction between the exploitation ability and the exploration ability.This paper presents a scheme that introduced main swarm to search for global solution and subswarm which was named as vanguard swarm to search for accurate local solution, a severance rule was introduced to re-disperse the particles of vanguard swarm after the local search.The Particle Swarm Optimization with this novel search schem was called Vanguard Scheme Particle Swarm Optimization (VPSO).A comparison between the VPSO and the other 4 variants of PSO on 7 test functions indicates that the proposed algorithm can balance the contradiction between the exploitation ability and the exploration ability, and possesses a more broad application prospects.