Perspectives on Machine Learning-augmented Reynolds-averaged and Large Eddy Simulation Models of Turbulence

Abstract

This work presents a review and perspectives on recent developments in the use of machine learning (ML) to augment Reynolds-averaged Navier--Stokes (RANS) and Large Eddy Simulation (LES) models of turbulent flows. Different approaches of applying supervised learning to represent unclosed terms, model discrepancies and sub-filter scales are discussed in the context of RANS and LES modeling. Particular emphasis is placed on the impact of the training procedure on the consistency of ML augmentations with the underlying physical model. Techniques to promote model-consistent training, and to avoid the requirement of full fields of direct numerical simulation data are detailed. This is followed by a discussion of physics-informed and mathematical considerations on the choice of the feature space, and imposition of constraints on the ML model. With a view towards developing generalizable ML-augmented RANS and LES models, outstanding challenges are discussed, and perspectives are provided. While the promise of ML-augmented turbulence modeling is clear, and successes have been demonstrated in isolated scenarios, a general consensus of this paper is that truly generalizable models require model-consistent training with careful characterization of underlying assumptions and imposition of physically and mathematically informed priors and constraints to account for the inevitable shortage of data relevant to predictions of interest. Thus, machine learning should be viewed as one tool in the turbulence modeler's toolkit. This modeling endeavor requires multi-disciplinary advances, and thus the target audience for this paper is the fluid mechanics community, as well as the computational science and machine learning communities.

References in corpus (8)

• Machine learning based spatio-temporal super resolution reconstruction of turbulent flows
• A neural network approach for the blind deconvolution of turbulent flows
• Formulating turbulence closures using sparse regression with embedded form invariance
• Deconvolutional artificial neural network models for large eddy simulation of turbulence
• Mean-flow Data Assimilation based on minimal correction of turbulence models: application to turbulent high-Reynolds number backward-facing step
• Turbulence closure modeling with data-driven techniques: physical compatibility and consistency considerations
• Non-Markovian Closure Models for Large Eddy Simulations using the Mori-Zwanzig Formalism
• A priori analysis on deep learning of subgrid-scale parameterizations for Kraichnan turbulence

Cited by in corpus (53)

• Convolutional neural networks for fluid flow analysis: toward effective metamodeling and low-dimensionalization
• Sparse identification of nonlinear dynamics with low-dimensionalized flow representations
• Ensemble Kalman method for learning turbulence models from indirect observation data
• Experimental velocity data estimation for imperfect particle images using machine learning
• Generalization techniques of neural networks for fluid flow estimation
• Machine learning building-block-flow wall model for large-eddy simulation
• A curated dataset for data-driven turbulence modelling
• 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
• CFD-driven Symbolic Identification of Algebraic Reynolds-Stress Models
• Assessments of epistemic uncertainty using Gaussian stochastic weight averaging for fluid-flow regression
• Deep Structured Neural Networks for Turbulence Closure Modelling
• Interpreting neural network models of residual scalar flux
• An iterative data-driven turbulence modeling framework based on Reynolds stress representation
• Influence of adversarial training on super-resolution turbulence reconstruction
• A projection-based, semi-implicit time-stepping approach for the Cahn-Hilliard Navier-Stokes equations on adaptive octree meshes
• Log-law recovery through reinforcement-learning wall model for large-eddy simulation
• Discovering explicit Reynolds-averaged turbulence closures for turbulent separated flows through deep learning-based symbolic regression with non-linear corrections
• Combining direct and indirect sparse data for learning generalizable turbulence models
• Learning Closed-form Equations for Subgrid-scale Closures from High-fidelity Data: Promises and Challenges
• Accelerating high order discontinuous Galerkin solvers using neural networks: 3D compressible Navier-Stokes equations
• Machine-learning accelerated turbulence modelling of transient flashing jets
• Discovery of interpretable structural model errors by combining Bayesian sparse regression and data assimilation: A chaotic Kuramoto-Sivashinsky test case
• Robust training approach of neural networks for fluid flow state estimations
• A physics-inspired alternative to spatial filtering for large-eddy simulations of turbulent flows
• Frame invariant neural network closures for Kraichnan turbulence
• Space-dependent Aggregation of Stochastic Data-driven Turbulence Models
• A wall model for separated flows: embedded learning to improve a posteriori performance
• Discretize first, filter next: learning divergence-consistent closure models for large-eddy simulation
• Progressive augmentation of Reynolds stress tensor models for secondary flow prediction by computational fluid dynamics driven surrogate optimisation
• Filtered Partial Differential Equations: a robust surrogate constraint in physics-informed deep learning framework
• Data-driven approach for modeling Reynolds stress tensor with invariance preservation
• Validation and parameterization of a novel physics-constrained neural dynamics model applied to turbulent fluid flow
• Review of Physics-based and Data-driven Multiscale Simulation Methods for Computational Fluid Dynamics and Nuclear Thermal Hydraulics
• Physics-informed neural network applied to surface-tension-driven liquid film flows
• Neural network-augmented eddy viscosity closures for turbulent premixed jet flames
• Adjoint-Based Enforcement of State Constraints in PDE Optimization Problems
• A unified understanding of scale-resolving simulations and near-wall modeling of turbulent flows using optimal finite element projections
• Optimization of Second-Order Transport Models for Transition-Continuum Flows
• A Priori Uncertainty Quantification of Reacting Turbulence Closure Models using Bayesian Neural Networks
• Artificial neural network approach for turbulence models: A local framework
• Model order reduction with neural networks: Application to laminar and turbulent flows
• Physics-Based Machine Learning Closures and Wall Models for Hypersonic Transition-Continuum Boundary Layer Predictions
• A Model-Constrained Tangent Slope Learning Approach for Dynamical Systems
• Machine-learning-based simulation of turbulent flows over periodic hills using a hybrid U-Net and Fourier neural operator framework
• Data-Driven RANS Closures Using a Relative Importance Term Analysis Based Classifier for 2D and 3D Separated Flows
• Rotational and Reflectional Equivariant Convolutional Neural Network for data-limited applications: Multiphase Flow demonstration
• Turbulence closure modeling with data-driven techniques: Investigation of generalizable deep neural networks
• Data-driven detached-eddy simulations based on explicit algebraic stress expressions for turbulent flows
• Improved pressure-gradient sensor for the prediction of separation onset in RANS models
• Addressing A Posteriori Performance Degradation in Neural Network Subgrid Stress Models
• Data-driven quantification of model-form uncertainty in Reynolds-averaged simulations of wind farms
• Evaluation of Analytical Turbulence Closures for Quasi-Geostrophic Ocean Flows with Coastal Boundaries