Application of Statistical Noise Generation and Radiation for Aeroacoustic Predictions in the Automotive Industry

Flow-induced noise is an important issue in the automotive industry. TNO in cooperation with AVL developed an efficient hybrid Computational AeroAcoustics (CAA) method, aimed at actual engineering applications, i.e. fast, robust, accurate and able to handle highly complex geometries. The hybrid method is based on standard Reynolds Averaged Navier-Stokes solution of the flow field under consideration, and combines an Statistical Noise Generation and Radiation (SNGR) technique to generate a time-accurate realization of the turbulence and a high order, quadrature-free discontinuous Galerkin - Linearized Euler Equation (LEE) solver to predict the propagation of the acoustic waves. Owing to the unstructured grid strategy the present approach is both particularly well suited for the complex geometries common to car design as well as sufficiently flexible in selecting acoustically interesting subdomains.

This paper demonstrates the capabilities of the CAA method in two cases on the noise induced by the flow around a car, and more specifically, by the A-pillar and rear-view mirror on the side window.