Noise, Vibration and Harshness (NVH) has become crucial design parameter for automotives which is undergoing enormous technological advancements due to vehicle electrification where bulky, noisy IC Engine which mask driveline and other noises are being replaced by high-speed electric motor as a prime mover. In such electrified drive systems, along with power transmitting gears system, high speed electric motor also acts as source for generating vibrations and noise at vehicle level which are perceptible by end users. The combined effect of the electromechanical sources of excitation is even more severe which directly affect powertrain and subsequently vehicle performance in field. So, electrified drive systems need to be designed for better NVH performance considering motor characteristics curve, speed-torque requirements of application and to diligently capture voice of customer directed towards comfort with low noise levels. In view of this, assessing NVH performance of electrified drive system right at the design stage is of paramount importance.
In this paper, multidisciplinary, combined approach to analytically predict NVH performance of electrified drive systems is being discussed in terms of vibration accelerations on housing and at mounting locations under influence of electric motor excitations and gear excitations. This helps to evaluate severity of excitation sources, system susceptibility for transferring vibrations and system response for different excitation sources. Frequency domain based predictive dynamic system model comprising of detailed sub-systems is being developed to evaluate component level contributions and subsequent system level impact. Motor orders are evaluated based on motor dynamic characteristics. Gear excitations are evaluated from gear transmission errors. System response is predicted under influence of electromechanical excitations using virtual accelerometers. The predictive model can be validated using physical testing by capturing acceleration data at defined accelerometer locations and analyzing corresponding sources. The proposed methodology can be extended to variety of electrified drive systems to develop respective predictive models. Actionable insights from predictive models become critical inputs to product designers and can be implemented at initial stage of product design to ensure design for NVH.