摘要
The issue of optimizing the dynamic parameters in Particle Swarm Optimizer (PSO) is addressed in this paper. An algorithm is designed which makes all particles originally endowed with a certain level energy, what here we define as EPSO (Energy Strategy PSO). During the iterative process of PSO algorithm, the Inertia Weight is updated according to the calculation of the particle's energy. The portion ratio of the current residual energy to the initial endowed energy is used as the parameter Inertia Weight which aims to update the particles' velocity efficiently. By the simulation in a graph theoritical and a functional optimization problem respectively, it could be easily found that the rate of convergence in EPSO is obviously increased.
The issue of optimizing the dynamic parameters in Particle Swarm Optimizer （PSO） is addressed in this paper. An algorithm is designed which makes all particles originally endowed with a certain level energy, what here we define as EPSO （Energy Strategy PSO）. During the iterative process of PSO algorithm, the Inertia Weight is updated according to the calculation of the particle＇s energy. The portion ratio of the current residual energy to the initial endowed energy is used as the parameter Inertia Weight which aims to update the particles＇ velocity efficiently. By the simulation in a graph theoritical and a functional optimization problem respectively, it could be easily found that the rate of convergence in EPSO is obviously increased.
基金
National Natural Science Foundation of China (No.50408034)
