Correlation of NVH Model for Extended Range Generator in Electric Vehicle

Electrification in the automotive industry has been steadily rising in popularity for many years, and with any technology there is always a desire to reduce development cost by efficiently iterating designs using accurate simulation models. In the case of rotating machinery and other devices that produce vibrations, an important physical behavior to simulate is Noise Vibration and Harshness (NVH). Efficient workflow to account for NVH was established at Schaeffler for eMotor design. Quantitative prediction is difficult to achieve and is occasionally intended only for faster iterations and trend prediction. A good validated qualitative simulation model would help achieve early NVH risk assessment based on the specified requirement and provide design direction and feasibility guidance across the design process to mitigate NVH concerns. This paper seeks to provide a general approach to validate the simulation model. The correlation methods used in this paper consist of a combination of alignment with structure-borne noise (SBN) and experimental modal analysis (EMA). It is ideal to validate component-level EMA before final assembly, but for more complex components this can be difficult. Many checks must be made for the test including validating magnet temperatures, sensor positions, and boundary conditions. For the permanent magnet synchronous motor (PMSM) drive unit in this case study, predictions of the SBN trend and response magnitude were improved.