Robust Deep Learning For Emulating Turbulent Viscosities
Résumé
From the simplest models to complex deep neural networks, modeling turbulence with machine learning techniques still offers multiple challenges. In this context, the present contribution proposes a robust strategy using patch-based training to learn turbulent viscosity from flow velocities, and demonstrates its efficient use on the Spalart-Allmaras turbulence model. Training datasets are generated for flow past twodimensional obstacles at high Reynolds numbers and used to train an auto-encoder type convolutional neural network with local patch inputs. Compared to a standard training technique, patch-based learning not only yields increased accuracy but also reduces the computational cost required for training.
Domaines
Modélisation et simulation
Origine : Fichiers éditeurs autorisés sur une archive ouverte