Moving from a 3D Axial Flux Machine Model to 2D Considering the Impact of End Leakage Flux
Paper in proceeding, 2024

Axial Flux Machines (AFMs) may have a potential of smaller axial length and higher power and torque density compared to the Radial Flux Machines (RFMs). However, the AFM needs to be modeled in Three-Dimensions (3D) yielding a large computation time when solving the Finite Element Method (FEM) model. A larger extent of computation time can be reduced by transforming the 3D problem to a Two-Dimensional (2D) problem. A single computation plane located at the center of the magnet in the 3D model is employed. This transformation leads to a loss of information due to the leakage flux in the axial direction, mainly around the rotor magnets and around the stator end windings. The purpose of this paper is to transform the 3D AFM model to a 2D model and quantify the consequences. An off-the-shelf outer stator inner rotor reference AFM is compared with an equivalent 3D FEM model. Several 3D and 2D models with varying core and magnet lengths were compared. A model with narrow core and magnet lengths was further investigated for five different model sizes ranging from full 3D to 2D. The results for the investigated machine type reveal that the rotor magnet leakage contributes to a dominant effect that necessitates a 3D model if a magnetically leading rotor core surrounds the magnet in the radial direction, and when the radial thickness of the magnets is small. Similarly, winding end leakage effects must be modeled in 3D when the radial thickness of the stator core is small relative to the coil end extent.

3D to 2D transformation

Finite element method modeling

Axial flux permanent magnet machine

Author

Vineetha Puttaraj

Chalmers, Electrical Engineering, Electric Power Engineering

Sonja Lundmark

Chalmers, Electrical Engineering, Electric Power Engineering

Torbjörn Thiringer

Chalmers, Electrical Engineering, Electric Power Engineering

2024 International Conference on Electrical Machines, ICEM 2024


9798350370607 (ISBN)

2024 International Conference on Electrical Machines, ICEM 2024
Torino, Italy,

Subject Categories

Other Electrical Engineering, Electronic Engineering, Information Engineering

DOI

10.1109/ICEM60801.2024.10700366

More information

Latest update

11/8/2024