Simulation Diagnostics Approach For Source Identification And Quantification In NVH Development Of Electric Motors

After decades of research and development in diagnosing and fine-tuning NVH characteristics of conventional internal combustion (IC) engines, the rapid electrification in powertrains has posed new challenges to the NVH engineers. Elimination of the IC engines drastically reduces powertrain borne noise levels but unmasks other existing noise sources like wind, road, ancillary devices and squeak & rattle. In addition, the new tonal sounds from electro-mechanical drive systems have low overall noise levels but have high frequency components with high annoyance rating. In summary, the electrification of powertrains has shifted powertrain NVH development from overall level to sound quality with different targets requiring several electro-mechanical solutions with innovative simulation and testing approaches. The purpose of this paper is to present a novel diagnostic approach to detect and quantify the prominent sources of electric motor borne noise and vibrations considering the multi-physics interaction between electric motor structural mechanical properties and electromagnetic phenomena. The Maxwell electromagnetic excitations are computed in radial and tangential directions and are applied on the stator structural mesh to compute the Equivalent Radiated Power (ERP) using vibroacoustic simulations. A diagnostic approach is established to investigate the individual and interactive effects of Maxwell forces and structural modal behaviour of the electric motor on the ERP response. This helps in identifying and prioritizing the motor design parameters and structural modif