Fuzzy clustering enhanced competitive swarm optimizer for balancing convergence and diversity in large-scale multiobjective optimization

Yanyan Tan, Xiaojie Li, Yukun Zhang, Wei Zheng, Huaxiang Zhang

Abstract: Highlights•Enhancement of Decision Variable Analysis: Fuzzy clustering is applied to decision variables to analyze their contribution to both convergence and diversity, resulting in more accurate grouping through a two-dimensional membership vector.•Novel Population Segmentation Method: A new method segments the population into winners and losers based on non-dominated sorting and crowding distance, creating four distinct sub-populations for more efficient optimization.•Targeted Sub-population Update Strategy: A new update strategy is proposed for the sub-populations, utilizing the membership degree vectors to optimize convergence or diversity.•Improved Performance on Benchmark LSMOPs: The proposed method outperforms five popular multi-objective optimization algorithms on nine benchmark LSMOPs, demonstrating its effectiveness.

External IDs:dblp:journals/eswa/TanLZZZ26