Broadband Noise Prediction of Stochastic Sources Based on the Linearized Euler Equations

Hybrid computational methods for aeroacoustics rely on the correct estimation of the flow solution to properly determine the equivalent noise sources. This flow solution may be obtained using unsteady CFD simulations that, in most cases, are prohibitive for industrial applications due to the large computational cost required. The hybrid approach for computational aeroacoustics presented in this work is based on SNGR method (Stochastic Noise Generation and Radiation) for synthetizing noise sources to be applied in the right-hand side of the Linearized Euler Equations (LEE). The stochastic method for generating the aeroacoustic sources depend on the time averaged solution of the flow field (RANS), reducing the computational cost associated to the CFD simulation. On the other hand, the acoustic propagation is solved by means of a high-order adaptive Discontinuous Galerkin (DG) scheme in time domain. This scheme is chosen due to its ability to accurately represent the phenomena involved and its high parallel scalability allowing the analysis of large/high frequency problems at acceptable computational costs from an industrial perspective. Subsequently, the hybrid method proposed is applied to a full-size car to compute the broadband acoustic field. Finally, the influence of the car's speed on the noise generation is assessed.