Research On Multi-objective Evolutionary Algorithm Based On Comprehensive Score And Convergence Ratio

The multi-objective optimization problem has important research value in the production and life of today's society.An evolutionary algorithm is one of the efficient algorithms to deal with this problem.The multi-objective evolutionary algorithm has evolved according to Darwin's biological evolution ideas.After introducing a variety of mechanisms,it is currently being widely studied by scholars from various countries and has achieved remarkable results.It is in a stage of rapid development.Among them,the multi-objective evolutionary algorithm based on Pareto is one of the research hotspots in recent years,and the overall performance of the algorithm is better.However,in the multi-objective evolutionary algorithm based on Pareto,due to the increase in the number of targets and the increase in the population base,a large number of solutions have obtained the same priority.Solutions with the same priority are difficult to distinguish during environmental selection,choosing algorithm The increased difficulty impedes the performance and efficiency of the algorithm.At the same time,the multi-objective evolutionary algorithm based on Pareto often uses the ranking selection method for the environment selection step,and the idea of this method is conducive to improving the convergence rate of non-inferior solutions,but it is difficult to protect the diversity of population genes,and it is impossible to balance convergence and diversity Promote.To solve the above problems,this text makes the following innovations and work:(1)Improve the ranking method and propose a comprehensive scoring parameter as the second selection criterion of the ranking method;(2)The improved ranking method is used in the multi-objective evolution algorithm based on Pareto,and the environment selection strategy is improved based on using the second selection criterion,and a new algorithm is proposed.A large number of horizontal and vertical comparison experiments were performed on the new algorithm to verify the excellent performance of the algorithm;(3)Based on the preference mechanism,a manual adjustment parameter,the convergence ratio,is proposed.The performance of the algorithm can be improved by adjusting the parameter to the actual problem of the decision-maker;(4)In recent years,the convergence ratio is quoted in the hotspot algorithm,and experiments on the improved algorithm are conducted to verify that this parameter can improve the overall performance of the algorithm.It can be summarized as this paper studies the multi-objective optimization algorithm based on Pareto.Based on the analysis of related algorithms,two improved multi-objective optimization algorithms based on comprehensive scoring and based on the convergence ratio are proposed,and through experiments and other algorithms The comparison proves the effectiveness of the algorithm in this text.