Artificial Intelligence-Driven Electric Motor Design: A Design-Task-Oriented Review

Artificial intelligence (AI) techniques are increasingly applied in electric motor design to meet the growing demands for high efficiency, high power density, and rapid design iteration. Traditional design approaches based on analytical modeling and iterative finite-element analysis face significant challenges when addressing high-dimensional design spaces and strongly coupled multi-physics constraints. This paper presents a comprehensive review of AI-driven motor design from a design-task-oriented and physics-aware perspective. Rather than organizing the literature by algorithms, the review systematically maps AI techniques to key motor design tasks, including electromagnetic optimization, thermal and mechanical design, multi-physics coupling, material selection, and manufacturability-aware optimization. Special emphasis is placed on AI-physics hybrid frameworks, such as surrogate-ch enhance model reliability and engineering interpretability. Practical challenges and emerging trends toward intelligent and automated motor design are also discussed, providing a structured reference for integrating AI into motor design.