Multi-objective hierarchical optimization method for pre-design of permanent magnet eddy current coupling

Le résumé fourni par la source

Abstract Obtaining the optimal geometric parameters for a physical device is always a complex and time-consuming process. Aiming to improve the synthetic performance and design efficiency for permanent magnet eddy current coupling (PMECC), a Kriging-assisted multi-objective hierarchical optimization (KMHO) method is proposed in this paper. For the first time, the sensitivity and coupling degree are jointly adopted to stratify the design geometric parameters through the clustering method. Then, the Kriging model is employed to simplify the construction process of the optimization objective function in each layer. Next, the Multi-Objective Genetic Algorithm (MOGA) is used to maximize the output torque and minimize the axial force, and the volume is chosen as the optimal selection basis. Ultimately, the application results indicate that the proposed method can achieve satisfactory design outcome, and significantly reduce the design time, because a single cycle is sufficient for convergence. In addition, further analysis indicates the optimization results also take on excellent performance in non-rated operating conditions. Therefore, the method will provide a reference framework for multi-objective optimization design of other similar physical devices and instruments.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.