Particle Swarm Optimization with Escape Velocity
Xiuli Wang, Yongji Wang, Haitao Zeng, Hui Xiang Zhou · 2006
This paper presents a model of Particle Swarm Optimization with Escape Velocity (EVPSO) in order to overcome premature convergence in the basic Particle Swarm Optimization (PSO). The EVPSO model equips particles with the escape velocity to avoid them trapping into local minima and increase the diversity of population. A simulation study shows that the EVPSO outperforms the basic PSO, especially for high dimension function. The EVPSO model facilitates solving multi-modal optimization problems.