Data-driven subgrid-scale modeling of forced Burgers turbulence using deep learning with generalization to higher Reynolds numbers via transfer learning

TL;DR

This paper demonstrates that deep learning-based subgrid-scale models for turbulence can be stabilized and generalized to higher Reynolds numbers through proper data conditioning and transfer learning, improving LES simulations.

Contribution

The study introduces a data-driven SGS modeling approach using deep neural networks with pre-conditioned data and transfer learning for better stability and generalization in turbulence simulations.

Findings

Deep neural networks provide stable LES models for Burgers turbulence.

Transfer learning enables accurate modeling at Reynolds numbers ten times higher.

Proper data augmentation improves model stability and extrapolation.

Abstract

Developing data-driven subgrid-scale (SGS) models for large eddy simulations (LES) has received substantial attention recently. Despite some success, particularly in a priori (offline) tests, challenges have been identified that include numerical instabilities in a posteriori (online) tests and generalization (i.e., extrapolation) of trained data-driven SGS models, for example to higher Reynolds numbers. Here, using the stochastically forced Burgers turbulence as the test-bed, we show that deep neural networks trained using properly pre-conditioned (augmented) data yield stable and accurate a posteriori LES models. Furthermore, we show that transfer learning enables accurate/stable generalization to a flow with 10x higher Reynolds number.