A Prediction Strategy Based On Decision Variable Analysis For Dynamic Multiobjective Evolutionary Optimization

When solving the multi-objective optimization problem,there are multiple conflicting targets and the optimization target will change with time,which is called dynamic multi-objective optimization problems(DMOPs).This kind of problem emphasizes the timeliness of the algorithm and requires the algorithm to maintain good diversity and convergence in the dynamic environment.Since classical multi-objective Optimization Algorithms(MOEAs)cannot detect the environment changes and lack a good diversity maintenance mechanism when solving such problems,such Algorithms cannot effectively solve DMOPs.In order to enable MOEAs to better solve DMOPs,it is necessary to introduce an environment detection mechanism,and increase the response mechanism after detecting changes in the environment,so as to improve the performance of the algorithm.In order to solve this problem,this paper proposes a strategy based on decision variable classification to respond to environmental changes.The strategy consists of three steps.After detecting the environment change,the first step is to analyze the influence of each decision variable on individual convergence and distribution in the new environment.The second step is to adopt different prediction methods for different decision variables.Finally,adaptive selection is applied to the solution set generated in the first and second steps,and solutions with good convergence and diversity are selected to make the initial population more adaptable to the new environment.The prediction strategy can help the solution set converge while maintaining its diversity.The experimental results and performance show that the proposed algorithm is capable of significantly improving the dynamic optimization performance compared with five state-of the-art evolutionary algorithms.