Wall model based on neural networks for LES of turbulent flows over periodic hills
Abstract
In this work, a data-driven wall model for turbulent flows over periodic hills is developed using the feedforward neural network (FNN) and data from wall-resolved large-eddy simulation (WRLES). To develop a wall model applicable to different flow regimes, the flow data in the near-wall region at all streamwise locations are grouped together as the training data set. In the developed FNN wall models, we employ the wall-normal distance, near-wall velocities, and pressure gradients as input features and the wall shear stresses as output labels, respectively. A priori tests on the prediction accuracy and generalization capacity of the trained FNN wall model are carried out by comparing the predicted wall shear stresses with the WRLES data from the same cases for model training and the cases with different Reynolds numbers and hill geometries. For the instantaneous wall shear stress, the FNN predictions show an overall good agreement with the WRLES data with some discrepancies observed at locations near the crest of the hill. The correlation coefficients between the FNN predictions and WRLES predictions are larger than 0.7 at most streamwise locations. For the mean wall shear stress, the FNN predictions agree very well with WRLES data. A posteriori test is also carried out. A good performance is observed for the turbulent channel flow case. Discrepancies between the predictions from the wall-modeled LES and the WRLES are observed for the periodic hill case.
- Publication:
-
Physical Review Fluids
- Pub Date:
- May 2021
- DOI:
- 10.1103/PhysRevFluids.6.054610
- arXiv:
- arXiv:2011.04157
- Bibcode:
- 2021PhRvF...6e4610Z
- Keywords:
-
- Physics - Fluid Dynamics