A New Particle Swarm Optimization Algorithm with Random Inertia Weight and Evolution Strategy
Gao Yue-lin, Duan Yu-hong · 2007 International Conference on Computational Intelligence and Security Workshops (CISW 2007) · 2007
The paper gives a new particle swarm optimization algorithm with random inertia weight and evolution strategy (REPSO). The proposed random inertia weight is using simulated annealing idea and the given evolution strategy is using the fitness variance of particles to improve the global search ability of PSO. The experiments with six benchmark functions show that the convergent speed and accuracy of REPSO is significantly superior to the one of The PSO with linearly decreasing inertia weight LDW-PSO.