Binary Group Adaptive Variation Multi-objective Particle Swarm Optimization Algorithm
Multi-objective particle swarm optimization has been widely concerned because of its powerful search ability and simple im-plementation,but it often has an imbalance between convergence and diversity.To solve this problem,a binary group adaptive variation multi-objective particle swarm optimization(BAMOPSO)algorithm is proposed to better balance convergence and diversity.Firstly,a novel binary group initialization method is used to make the particles more evenly distributed in the target space,thus improving the di-versity of the algorithm.Secondly,a new adaptive variation strategy is designed to improve the convergence according to the change of the objective function value during evolution.Finally,by comparing BAMOPSO algorithm with six classical multi-objective optimiza-tion algorithms on 15 benchmark functions,it is verified that the algorithm has a significant improvement in balance convergence and diversity.
万方数据
维普
