Design Technique of Open-Slot Surface-Mounted PM Machines Using Coenergy-Based Subdomain Model and Teaching-Learning-Based Optimization Algorithm

The open-slot stator topology, which allows for preformed coils and an excellent fill factor, has been regarded as a suitable candidate for low-speed operation such as direct-drive systems. This paper presents a rapid and accurate design optimization of surface-mounted permanent magnet (PM) machines equipped with the open-slot stator configuration. The analytical method, utilizing the coenergy-based subdomain model, is developed to predict the various electromagnetic performances. The proposed analytical model is decoupled into two components, relevant and irrelevant to the optimization process. Subsequently, the teaching-learning-based optimization (TLBO), a human-based algorithm for global solutions, is applied to the analytical model to search for the best compromise belonging to the Pareto-optimal front. The design methodology presented in this article is well-suited for the open-slot stator configuration, providing a computationally efficient and accurate approach to address multi-constrained minimization problems. In the end, the reliability of the proposed model is verified by finite element (FE) results.