A new two-stage particle swarm optimization algorithm

As inertia weight is an important parameter to balance global research and local research, a two-stage particle swarm optimization algorithm was proposed. In the first stage, the algorithm used dynamic and self-adapting inertia weight based on different dimensions and different particle to accelerate the convergent speed; in the second stage, it used linear inertia weight and chaotic mutation to prevent local optimum. At last, experimental results for seven typical test show that this algorithm(2-SPSO) is better than PSO and LDIWPSO in speed, precision of convergence and capacity of global optimization.