An Efficient Hybrid Computational Process for Aeroacoustic Vehicle Development Based on Navier-Stokes Equations and Lighthill's Acoustic Analogy

Vehicle interior noise at highway speeds is dominated by aerodynamic sources arising from unsteady turbulent flow around external components. Carlo Alberto Perugini presents a hybrid computational methodology for predicting aerodynamic noise sources and their contribution to interior cabin noise in road vehicles. His approach combines unsteady CFD simulations based on the incompressible Navier-Stokes equations with a finite element acoustic model derived from Lighthill's acoustic analogy and a vibro-acoustic model for noise propagation into the cabin. The CFD simulations are performed using a DDES approach to resolve large-scale turbulent structures, while the acoustic propagation is computed using FEM. The resulting pressure fluctuations are used as input for the vibro-acoustic model to evaluate interior noise levels. The methodology is applied to a full-scale production SUV, with a detailed aeroacoustic analysis of the side mirror and roof spoiler regions. Experimental measurements conducted in an aeroacoustic wind tunnel are used for validation, including velocity field data, exterior sound pressure levels, and interior cabin noise measurements. The results show good agreement between numerical predictions and experimental data across the investigated frequency range. The study highlights the contribution of specific flow structures to noise generation and provides a structured and reproducible workflow for aeroacoustic analysis of vehicle components.