A novel adaptive coevolution based multi-objective robust optimization strategy

Multi-objective optimization addresses engineering design problems with multiple objectives, often encountered in practical engineering, which typically involves various sources of uncertainty that can significantly impact design performance. To account for these uncertainties, robust design methodologies have been developed to achieve solutions that are both optimal and resilient to uncertainty. However, performing robustness analysis in the context of robust design, particularly when consider ing interval uncertainties, can be extremely time-consuming. To address this issue, a new algorithm is proposed in this paper, which is named Adaptive Coevolution-Based Multi-Objective Robust Optimization (AC-MORO). This method employs an adaptive approach to determine the neighborhoods between the robustness analysis problems associated with candidate solutions. In contrast to the conventional approach of analyzing the robustness of a single candidate solution at a time, the proposed meth od performs a simultaneous robustness analysis of multiple candidate solutions, utilizing the information obtained from their respective neighborhoods. To demonstrate the efficacy of the proposed algorithm, a series of tests and evaluations were conducted on six benchmark problems.