Comparison of analytical models for predicting cogging torque in surface-mounted PM machines

The paper compares five analytical models for predicting the cogging torque in surface-mounted permanent magnet machines, viz. lateral force (LF), complex permeance (CP), and combined CP models, together with subdomain (SD) models based on single slot/pole and exact SD model. The models are evaluated for the cogging torque waveforms and the prediction of optimal pole-arc to pole-pitch and slot-opening to slot-pitch ratios for cogging torque reduction. It shows that all analytical models predict similar waveforms but different magnitudes. The performance of the LF model is very stable and it is very accurate for slot-opening to slot-pitch ratio less than 0.2, but tends to underestimate above this. Both CP models have similar and relatively lower accuracy and usually overestimate the cogging torque due to neglecting the deformations of magnets and the path to predict the airgap flux density in the conformal mapping. The SD models have high accuracy while the exact SD model is most accurate by accounting for mutual influence between slots.