Optimisation of Rotor Skewing of a Yokeless and Segmented Armature Axial Flux Machine

This paper presents an optimisation approach for rotor skewing in a Yokeless and Segmented Armature (YASA) design Axial Flux Machine (AFM) for electric vehicle applications. Torque ripple amplitudes are a critical factor influencing the noise, vibration and harshness (NVH) behaviour of electric motors. The focus of this paper is to reduce the torque ripple amplitudes of the dominant harmonics over the entire torque-speed characteristic of the AFM. The principle of the proposed approach is a segmented permanent magnet configuration of the AFM, where individual magnet segments can be circumferentially shifted to achieve optimal skewing configurations. Initial optimisations are performed using 2D finite element (FE) simulations, modelled as linear motors with multiple slices and different numbers of magnet segmentation. However, the accuracy of the 2D FE results is limited due to the lack of interaction between the individual segments and the insufficient representation of three-dimensional (3D) effects. To overcome these limitations, the optimisation is extended to 3D FE simulations. This 3D optimisation, conducted with five and 20 magnet segments, demonstrates a significant reduction of torque ripple amplitudes across the entire operating range while adhering to predefined constraints. Finally, different optimised skewing configurations are evaluated considering demagnetisation in 3D FE simulations.