Particle Swarm Optimization Based on Number-theoretic Method
基于数论方法的微粒群算法

An improved particle swarm optimization algorithm is proposed. Number-theoretic method is used to initialize the particles' position. The particle locates on the current global minimum is re-initialized during the algorithm. In the end the result of PSO is fined by a method named creeping algorithm. The experiments manifest that the improved algorithm can get the better results.