An Evolutionary Algorithm with Feasibility Tracking Strategy for Constrained Multi-Objective Optimization Problems

In recent years, many constrained multi-objective evolutionary algorithms have been proposed to address com-plex constrained multi-objective optimization problems and have shown significant performance. However, when facing challenging constrained multi-objective optimization problem, some algorithms struggle to explore all feasible regions. To address this issue, this paper proposes a feasibility tracking strategy. By setting an expansion value and controlling constraint relaxation, adaptive adjustment of the reference vectors simulates an ex-panded narrow feasible region. After the preliminary exploration of the feasible region, a new set of reference vectors is generated to ensure the existence of feasible regions in the direction of the new reference vectors, thereby increasing the density of individuals within the feasible region. Experimental results on a series of benchmark problems demonstrate that the proposed algorithm is quite effective compared to other state-of-the-art constrained multi-objective evolutionary algorithms.