Design and Cogging Torque Optimization of Axial-Radial Flux Permanent Magnet Machine With Novel Modular Stator

This paper investigates the design and optimization of an axial-radial flux permanent magnet (ARFPM) machine, which is an attractive option for high torque density applications. First, considering manufacturing costs and prevention of axial displacement, a novel modular stator is proposed. The struc...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on industry applications Vol. 61; no. 5; pp. 7023 - 7034
Main Authors Long, Yuhang, Song, Zhanfeng, Zhang, Zan
Format Journal Article
LanguageEnglish
Published IEEE 01.09.2025
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:This paper investigates the design and optimization of an axial-radial flux permanent magnet (ARFPM) machine, which is an attractive option for high torque density applications. First, considering manufacturing costs and prevention of axial displacement, a novel modular stator is proposed. The structural design, assembly, and fixation of which are revealed in detail, and the mechanical strength is analyzed to validate its feasibility. The coupling between the axial flux and radial flux is studied by equivalent magnetic circuit (EMC) and three-dimensional (3-D) finite element analysis (FEA). The design of key parameters to improve torque density is also discussed. Then, a comprehensive comparison of the electromagnetic performance is performed among three types of ARFPM machines: the one with a conventional soft magnetic composite (SMC) stator manufactured by molding, the one with the proposed modular stator, and the one with a traditional integrated silicon steel stator. In order to address the common issue of cogging torque without any additional material costs, the combination of PMs with non-uniform pole arcs is suggested and optimized. Finally, this research is one of the few that manufacture prototypes of 3-D hybrid flux PM machines. The effectiveness of analytical methods and the accuracy of simulated results are validated by the tests on two ARFPM machines before and after improvement.
ISSN:0093-9994
1939-9367
DOI:10.1109/TIA.2025.3556801