Development of an Aeroacoustic Simulation Procedure for Passenger Vehicles: An investigation on autonomous vehicle wind noise

Abstract

With the shift to electric powertrains, wind noise has become a proportionally larger contributor to the overall noise levels inside the vehicle. Considering the correlation between noise and premium brand quality and the demand by the customers for quieter vehicles, manufacturers are spending more resources determining noise contribution and ways to quantify, understand and dampen sources of wind noise. The cost of performing aeroacoustic wind tunnel experiments is high. Therefore, methods to predict wind noise using aeroacoustic computational fluid dynamics (CFD) simulations can establish a correlation between wind noise outside and inside the cabin at an earlier stage in the product development. Experimental data was obtained from accelerometers placed on the left-hand side driver’s window and the quarter glass window. An aeroacoustic CFD simulation method was developed and a correlation was obtained with the experimental data. The pressure fluctuations were simulated using unsteady scale resolving flow simulations, with the data being stored as a time series data. A fast fourier transform (FFT) was implemented to decompose the pressure data into its constituent frequency spectrum. Both simulation and experimental data were analyzed using third octave analysis to determine the sound pressure level (SPL) for discretized octave bands. Studies for the mesh size and the time step were carried out to ensure the pressure waves are sampled adequately spatially and in time. Comparisons were made by means of a delta analysis between baseline and modified configurations with results showing that the simulation was accurate in predicting SPL delta values and trends for most configurations. The simulation method was implemented on a range of geometry changes pertaining to the placement of the advanced driver assistance systems (ADAS) to evaluate the change in the noise levels on the external body due to the addition of driver assistance systems.