Hybridized Stator Characterization Technique to Accurately Estimate the Peak Performance of Automotive Electric Machines

The rapid expansion of the production of electric vehicles (EV) has increased the demand for high-performance electric motors. The B-H and iron loss of the stator core significantly influence the resulting torque and efficiency of the electric machine. Manufacturing is a major factor that affect the B-H and iron losses of the stator core. Therefore, precise magnetic core characterisation is essential for ensuring reliability and efficiency. End-of-line (EoL) testing a critical validation step, assessing product integrity, defect identification, and regulatory compliance. However, accurate characterisation of the magnetic properties of complex geometries, such as stator cores, is not feasible using standardised measurement methods, which are typically designed for simpler shapes such as toroidal samples. In this paper a hybrid magnetic characterisation method has been developed in which standard experimental data is combined with finite element analysis to improve the estimation of the magnetic characteristics of a stator core material. By compensating for the limitations of conventional stator testing, the proposed method achieves a substantial improvement in material characterisation accuracy by around 69% for the hybrid electric vehicle (HEV) machine and 75% for the electric vehicle (EV) machine. When these refined material properties are employed in finite element motor models, the resulting torque prediction error is reduced by 94.43% in the HEV motor and by 98.39% in the EV motor, relative to predictions based on conventional stator sample testing. The improved agreement between predicted and expected performance demonstrates the relevance of the proposed approach for reliable motor design, particularly in the context of high-volume automotive manufacturing where consistency and repeatability are critical.