In the current scenario of EV revolution in the automotive industry, NVH
performance of the vehicles is one of the major points of sale to the customers.
Auxiliary components play one of the predominant roles in the contribution of
noise to overall vehicle interior or exterior sound pressure levels, which
impact customer vehicle comfort. CAE prediction of NVH performance of automotive
components involves a lot of design iterative processes, large server space
utilization, and time-consuming.
To reduce cost and time, data-driven technologies like AI algorithms can help CAE
engineers because of their high efficiency and high precision. In the current
research, a wiper motor mount stiffness prediction algorithm was designed based
on the historical data using CAE analysis and AI algorithms, and improved
prediction accuracy by tuning the parameters of AI algorithms using grid search
methodology.
High prediction accuracy of wiper motor mount stiffness has been achieved with
the method of support vector machine. CAE engineers can avoid iterative
processes by utilizing the optimized design parameters from the prediction
results without running full finite element analysis simulations.