Adaptive Weight Particle Swarm Optimization Algorithm with Constriction Factor

In order to overcome the shortage of premature convergence caused by local optimization in the process of global optimization, an adaptive weight Particle Swarm Optimization algorithm with constriction factor is proposed combined with an analysis of convergence of Particle Swarm Optimization algorithm. The value of the inertia weight is set according to dynamic information about the changes in the objective function value, as to effectively balance the advantages of global optimization against the shortage of local optimization. Four Benchmark function are used for performance test of five different kinds of optimization algorithm, the final results shows that the proposed method has a good ability to slow down the pace of premature convergence, compared to other improved particle swarm algorithm.