Apriori grid cut-off frequency estimation for Aeroacoustic simulations
* Presenting author
Abstract:
Aeroacoustic simulations require instationary flow field data. In most cases, flow simulations are significantly more expensive than sound propagation simulations. In general, the maximum frequency of interest in the sound spectra is known prior to the simulation. Thus, the flow field is often modelled by scale-resolving simulations whereby turbulence is only partially captured by the grid depending on the local cell-size. On one hand, the highest resolvable frequency and the simulation accuracy is inversely related to the cell-size, on the other hand, the computational cost scales at least linearly with the number of grid-cells. This trade-off between the accuracy and the computational complexity calls for a systematic approach for determining the local grid-resolution of the flow domain. Therefore, the motivation of this work is to establish a clear relation between the maximum resolved frequency in the sound spectra and the required local grid-resolution. For this purpose, we investigate common approaches in the literature for estimating the grid cut-off frequency in flow simulations. Besides, we propose a new approach based on the eddy turnover time and Kolmogov’s energy law for the inertial subrange. Lastly, we compare the results of different approaches in our test-cases.