An Efficient Framework Towards the Prediction of Trailing Edge Noise
* Presenting author
Abstract:
Hybrid simulation methods are often used in computational aeroacoustics. Here, the sources for the acoustic wave propagation are extracted from the flow data. Low-fidelity models such as RANS hydrodynamic simulations can reach their limits in terms of accuracy. Better results are obtained by high-fidelity models, such as wall-resolved LES. To close the gap between computational effort and accuracy, one approach is to restrict the LES domain to the region, in which the acoustic source terms have physical relevance. However, this requires a method to generate the inflow turbulence in the LES region. Here, the recently introduced RRALF method is used, which relies on time-averaged flow statistics to generate turbulence. For generic problems, pre-calculated data from a database can be used. But complex flow cases require a different approach. In this talk, an efficient methodology is presented to model the Reynolds stresses based on the distribution of the turbulent kinetic energy, obtained from a two-dimensional RANS simulation. This approach is possible as RANS promises to accurately predict the flow field outside critical regions, such as flow separation. Results of a zonal trailing edge simulation of an airfoil are presented and will be used to derive acoustic sources in a next step.