A MOEA/D Framework Based on Memory Differential Cooperation Mechanism for Dynamic Multiobjective Optimization

In this paper, a new framework based on memory differential cooperation mechanism is proposed. It effectively combines the memory mechanism, differential evolution algorithm and cooperation mechanism, and makes improvements according to the characteristics of MOEA/D to promote MOEA/D make full use of the historical evolution direction and PS distribution. Firstly, this paper designs a memory mechanism method to record the past historical information effectively, and selects the excellent individuals in the historical information as the guide group. Secondly, the differential cooperation mechanism will guide the population to decompose each subpopulation, and use the differential algorithm to iteratively optimize the subpopulation to search the Pareto set suitable for the current environment. Thirdly, through the idea of cooperation algorithm, the excellent individuals of each subpopulation are further shared to replace the individuals of this population, which further improves the convergence and diversity of the algorithm. Finally, extensive testing is conducted on FDA, dMOP, and F test suites. Experimental results show that exceeds the performance of some state-of-the-art algorithms.