A new constrained multi-objective optimization problems algorithm based on group-sorting
Yijun Liu, Xin Li, Qijia Hao · Proceedings of the Genetic and Evolutionary Computation Conference Companion · 2019
Constrained multi-objective optimization problems(CMOPs) have wide applications in many areas. One of the difficulties in solving CMOPs is to handle constraints and optimize objective values simultaneously. In this paper, three improvements are proposed for CMOPs. Firstly, a new strategy of adaptive grouping is proposed to ensure the diversity of the algorithm. Secondly, each sub-population evolves according to the designed crossover operator, which improves the searchability of the algorithm, thus accelerates the convergence process of the algorithm. Finally, a new sorting method based on the information of representative solutions is designed to measure the quality of individuals. The experimental results show that the proposed algorithm performs better in convergence and diversity.