Computational Prediction of Vehicle Aerodynamic Noise by Integration of a CFD Technique with Lighthill's Acoustic Analogy

Objective of the present work is the development and validation of a computational methodology for the prediction of flow-induced noise around a vehicle. The technique employed makes use of the Lighthill's acoustic analogy revised for low Mach number flows and consists of two steps: prediction of noise sources over the body surface by means of a suitable Computational Fluid Dynamics (CFD) technique and subsequent prediction of the radiated noise by solving the Ffowks-Williams Hawkings equation. For testing the methodology flows around a cylinder and a test shape, generating the type of flows occuring around the A-pillar of a car, have been employed. Results are compared with experimental data. The paper demonstrates the feasibility of computational prediction of aerodynamic noise even for turbulent high Reynolds number flows.