摘要:Sub-dimension particle swarm optimization(s-dPSO) is proposed based on basic particle swarm optimization (bPSO). Each dimension of particle in s-dPSO is updated in turn. The dimensions with poor diversity would be mutated that is initialized again to improve the diversity of population and get global optimal solution when the algorithm is in the local convergence. Most Benchmark function get good result with s-dPSO which ability of optimization is better than bPSO.