Physical invariance in neural networks for subgrid-scale scalar flux modeling

Abstract

In this paper we present a new strategy to model the subgrid-scale scalar flux in a three-dimensional turbulent incompressible flow using physics-informed neural networks (NNs). When trained from direct numerical simulation (DNS) data, state-of-the-art neural networks, such as convolutional neural networks, may not preserve well known physical priors, which may in turn question their application to real case-studies. To address this issue, we investigate hard and soft constraints into the model based on classical transformation invariances and symmetries derived from physical laws. From simulation-based experiments, we show that the proposed transformation-invariant NN model outperforms both purely data-driven ones as well as parametric state-of-the-art subgrid-scale models. The considered invariances are regarded as regularizers on physical metrics during the a priori evaluation and constrain the distribution tails of the predicted subgrid-scale term to be closer to the DNS. They also increase the stability and performance of the model when used as a surrogate during a large-eddy simulation. Moreover, the transformation-invariant NN is shown to generalize to regimes that have not been seen during the training phase.

arXiv comment32 pages, 17 figures, accepted for publication in Physical Review Fluids

References in corpus (10)

• Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation
• Turbulence Modeling in the Age of Data
• Deep Neural Networks for Data-Driven Turbulence Models
• Sub-grid modelling for two-dimensional turbulence using neural networks
• Searching for turbulence models by artificial neural network
• Training convolutional neural networks to estimate turbulent sub-grid scale reaction rates
• A priori analysis on deep learning of subgrid-scale parameterizations for Kraichnan turbulence
• Interpreting neural network models of residual scalar flux
• A posteriori analysis of low-pass spatial filters for approximate deconvolution large eddy simulations of homogeneous incompressible flows
• Analysis of the dissipative range of the energy spectrum in grid turbulence and in direct numerical simulations

Cited by in corpus (13)

• Enhancing Computational Fluid Dynamics with Machine Learning
• Learned discretizations for passive scalar advection in a 2-D turbulent flow
• Data-driven subgrid-scale modeling of forced Burgers turbulence using deep learning with generalization to higher Reynolds numbers via transfer learning
• A posteriori learning for quasi-geostrophic turbulence parametrization
• Learning physics-constrained subgrid-scale closures in the small-data regime for stable and accurate LES
• Deep reinforcement learning for large-eddy simulation modeling in wall-bounded turbulence
• Explaining the physics of transfer learning a data-driven subgrid-scale closure to a different turbulent flow
• Generative data-driven approaches for stochastic subgrid parameterizations in an idealized ocean model
• Investigation of nonlocal data-driven methods for subgrid-scale stress modelling in large eddy simulation
• Frame invariant neural network closures for Kraichnan turbulence
• Neural stochastic differential equations for particle dispersion in large-eddy simulations of homogeneous isotropic turbulence
• Bounded nonlinear forecasts of partially observed geophysical systems with physics-constrained deep learning
• Kinetic data-driven approach to turbulence subgrid modeling