Design Optimization and Scaling of Coreless AFPM Machines Using Hybrid FEA-Based Differential Evolution and Machine Learning

Authors

• Matin Vatani, University of KentuckyFollow
• David R. Stewart, University of KentuckyFollow
• Donovin D. Lewis, University of KentuckyFollow
• Dan M. Ionel, University of KentuckyFollow

Abstract

This paper presents a machine learning (ML) based design framework for the fast and accurate optimization of coreless axial flux permanent magnet (AFPM) machines. Although the absence of magnetic cores eliminates material nonlinearity, the design process remains highly nonlinear due to the complex influence of geometric parameters. To overcome the computational challenges of finite element analysis (FEA)-based optimization, a series of multi-objective differential evolution (MODE) optimizations were conducted across various machine sizes at constant power output. The resulting design data was used to train an artificial neural network (ANN), enabling rapid prediction of machine performance without the need for repeated FEA simulations. Case study results indicate that values of torque constant can be predicted with satisfactory accuracy, even for designs of a different size from the simulated set used for training. The proposed ML framework can significantly reduce design time and computational burden, providing a practical tool for efficiently exploring the design space in the optimization process.

Document Type

Conference Proceeding

Publication Date

10-2025

Digital Object Identifier (DOI)

doi: 10.1109/ECCE58356.2025.11259690

Repository Citation

Vatani, Matin; Stewart, David R.; Lewis, Donovin D.; and Ionel, Dan M., "Design Optimization and Scaling of Coreless AFPM Machines Using Hybrid FEA-Based Differential Evolution and Machine Learning" (2025). Electrical and Computer Engineering Graduate Research. 26.
https://uknowledge.uky.edu/ece_gradpub/26