Electromagnetic Analysis and Multi-Objective Design Optimization of a WFSM with Hybrid GOES-NOES Core

This study presents a design and optimization methodology to enhance the power density and efficiency of wound field synchronous machines (WFSMs) by selectively applying grain-oriented electrical steel (GOES). Unlike conventional non-grain-oriented electrical steel (NOES), GOES exhibits significantl...

Full description

Saved in:
Bibliographic Details
Main Authors: Kyeong-Tae Yu, Hwi-Rang Ban, Seong-Won Kim, Jun-Beom Park, Jang-Young Choi, Kyung-Hun Shin
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:World Electric Vehicle Journal
Subjects:
Online Access:https://www.mdpi.com/2032-6653/16/7/399
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study presents a design and optimization methodology to enhance the power density and efficiency of wound field synchronous machines (WFSMs) by selectively applying grain-oriented electrical steel (GOES). Unlike conventional non-grain-oriented electrical steel (NOES), GOES exhibits significantly lower core loss along its rolling direction, making it suitable for regions with predominantly alternating magnetic fields. Based on magnetic field analysis, four machine configurations were investigated, differing in the placement of GOES within stator and rotor teeth. Finite element analysis (FEA) was employed to compare electromagnetic performance across the configurations. Subsequently, a multi-objective optimization was conducted using Latin Hypercube Sampling, meta-modeling, and a genetic algorithm to maximize power density and efficiency while minimizing torque ripple. The optimized WFSM achieved a 13.97% increase in power density and a 1.0% improvement in efficiency compared to the baseline NOES model. These results demonstrate the feasibility of applying GOES in rotating machines to reduce core loss and improve overall performance, offering a viable alternative to rare-earth permanent magnet machines in xEV applications.
ISSN:2032-6653