Subgrid modelling for two-dimensional turbulence using neural networks
From MaRDI portal
Abstract: In this investigation, a data-driven turbulence closure framework is introduced and deployed for the sub-grid modelling of Kraichnan turbulence. The novelty of the proposed method lies in the fact that snapshots from high-fidelity numerical data are used to inform artificial neural networks for predicting the turbulence source term through localized grid-resolved information. In particular, our proposed methodology successfully establishes a map between inputs given by stencils of the vorticity and the streamfunction along with information from two well-known eddy-viscosity kernels. Through this we predict the sub-grid vorticity forcing in a temporally and spatially dynamic fashion. Our study is both a-priori and a-posteriori in nature. In the former, we present an extensive hyper-parameter optimization analysis in addition to learning quantification through probability density function based validation of sub-grid predictions. In the latter, we analyse the performance of our framework for flow evolution in a classical decaying two-dimensional turbulence test case in the presence of errors related to temporal and spatial discretization. Statistical assessments in the form of angle-averaged kinetic energy spectra demonstrate the promise of the proposed methodology for sub-grid quantity inference. In addition, it is also observed that some measure of a-posteriori error must be considered during optimal model selection for greater accuracy. The results in this article thus represent a promising development in the formalization of a framework for generation of heuristic-free turbulence closures from data.
Recommendations
- Sub-grid scale model classification and blending through deep learning
- Neural networks based subgrid scale modeling in large eddy simulations
- Invariant data-driven subgrid stress modeling in the strain-rate eigenframe for large eddy simulation
- A priori sub-grid modelling using artificial neural networks
- A neural network approach for the blind deconvolution of turbulent flows
Cites work
- A dynamic LES scheme for the vorticity transport equation: Formulation and a priori tests
- A dynamic localization model for large-eddy simulation of turbulent flows
- A dynamic subgrid-scale eddy viscosity model
- A neural network approach for the blind deconvolution of turbulent flows
- A paradigm for data-driven predictive modeling using field inversion and machine learning
- A stable and scale-aware dynamic modeling framework for subgrid-scale parameterizations of two-dimensional turbulence
- An eddy-viscosity subgrid-scale model for turbulent shear flow: Algebraic theory and applications
- Computational design for long-term numerical integration of the equations of fluid motion: two-dimensional incompressible flow. Part I
- Deep learning in fluid dynamics
- Hidden physics models: machine learning of nonlinear partial differential equations
- High-order methods for decaying two-dimensional homogeneous isotropic turbulence
- Introduction to geophysical fluid dynamics. Physical and numerical aspects
- Large Eddy Simulation for Incompressible Flows
- Learning partial differential equations via data discovery and sparse optimization
- Mathematics of Large Eddy Simulation of Turbulent Flows
- Neural network closures for nonlinear model order reduction
- Neural network modeling for near wall turbulent flow.
- Neural networks based subgrid scale modeling in large eddy simulations
- Optimal LES formulations for isotropic turbulence
- Quantifying and reducing model-form uncertainties in Reynolds-averaged Navier-Stokes simulations: a data-driven, physics-informed Bayesian approach
- Reynolds averaged turbulence modelling using deep neural networks with embedded invariance
- Smagorinsky constant in LES modeling of anisotropic MHD turbulence
- Subgrid modelling for geophysical flows
- Subgrid-scale backscatter in turbulent and transitional flows
- Theoretically based optimal large-eddy simulation
Cited in
(62)- Neural network modeling for near wall turbulent flow.
- A nonintrusive hybrid neural-physics modeling of incomplete dynamical systems: Lorenz equations
- Resampling with neural networks for stochastic parameterization in multiscale systems
- Assessments of epistemic uncertainty using Gaussian stochastic weight averaging for fluid-flow regression
- Data-driven multi-grid solver for accelerated pressure projection
- Frame invariant neural network closures for Kraichnan turbulence
- Spatiotemporally dynamic implicit large eddy simulation using machine learning classifiers
- Bridging the gap: machine learning to resolve improperly modeled dynamics
- Deep learning of the spanwise-averaged Navier-Stokes equations
- Simple computational strategies for more effective physics-informed neural networks modeling of turbulent natural convection
- Explicit physics-informed neural networks for nonlinear closure: the case of transport in tissues
- Stable \textit{a posteriori} LES of 2D turbulence using convolutional neural networks: backscattering analysis and generalization to higher \(Re\) via transfer learning
- Reducing data-driven dynamical subgrid scale models by physical constraints
- Testing of kinetic energy backscatter parameterizations in the NEMO ocean model
- Turbulence theories and statistical closure approaches
- A machine learning framework for LES closure terms
- Physics-informed regularization and structure preservation for learning stable reduced models from data with operator inference
- A data-driven approach to viscous fluid mechanics: the stationary case
- Prediction of turbulent heat transfer using convolutional neural networks
- A priori sub-grid modelling using artificial neural networks
- Learned turbulence modelling with differentiable fluid solvers: physics-based loss functions and optimisation horizons
- Modern Koopman theory for dynamical systems
- Interpreting neural network models of residual scalar flux
- Machine-learning-based spatio-temporal super resolution reconstruction of turbulent flows
- Unsupervised deep learning for super-resolution reconstruction of turbulence
- Reynolds-averaged Navier–Stokes equations with explicit data-driven Reynolds stress closure can be ill-conditioned
- Sub-grid scale model classification and blending through deep learning
- Neural network-based modelling of subsonic cavity flows
- Toward neural-network-based large eddy simulation: application to turbulent channel flow
- On the role of nonlinear correlations in reduced-order modelling
- On the Generalizability of Machine-Learning-Assisted Anisotropy Mappings for Predictive Turbulence Modelling
- Artificial-neural-network-based nonlinear algebraic models for large-eddy simulation of compressible wall-bounded turbulence
- Hybrid analysis and modeling, eclecticism, and multifidelity computing toward digital twin revolution
- Data-driven spectral turbulence modelling for Rayleigh–Bénard convection
- Parsimony as the ultimate regularizer for physics-informed machine learning
- A highly accurate strategy for data-driven turbulence modeling
- Forward sensitivity analysis and mode dependent control for closure modeling of Galerkin systems
- A data-driven method for modelling dissipation rates in stratified turbulence
- A data-driven statistical-stochastic surrogate modeling strategy for complex nonlinear non-stationary dynamics
- Learning subgrid-scale models with neural ordinary differential equations
- Differentiable hybrid neural modeling for fluid-structure interaction
- Energy-conserving neural network for turbulence closure modeling
- Bayesian conditional diffusion models for versatile spatiotemporal turbulence generation
- Discovery of interpretable structural model errors by combining Bayesian sparse regression and data assimilation: a chaotic Kuramoto-Sivashinsky test case
- Learning about structural errors in models of complex dynamical systems
- Deep learning in computational mechanics: a review
- Filtered partial differential equations: a robust surrogate constraint in physics-informed deep learning framework
- A transformer-based convolutional method to model inverse cascade in forced two-dimensional turbulence
- A recursive neural-network-based subgrid-scale model for large eddy simulation: application to homogeneous isotropic turbulence
- Neural differentiable modeling with diffusion-based super-resolution for two-dimensional spatiotemporal turbulence
- A wall model for separated flows: embedded learning to improve \textit{a posteriori} performance
- Modular operator superposition (MOS): a physics-guided machine learning framework for addressing the curse of dimensionality and multiscale challenges in computational fluid dynamics
- The discrete direct deconvolution model in the large eddy simulation of turbulence
- Data-driven stochastic closure modeling via conditional diffusion model and neural operator
- Physics informed neural network framework for unsteady discretized reduced order system
- Reduced data-driven turbulence closure for capturing long-term statistics
- Mitigating distribution shift in machine learning-augmented hybrid simulation
- Deconvolutional artificial-neural-network framework for subfilter-scale models of compressible turbulence
- Scientific machine learning for closure models in multiscale problems: a review
- Binned spectral power loss for improved prediction of chaotic systems
- Stochastic and non-local closure modeling for nonlinear dynamical systems via latent score-based generative models
- Dynamic mode decomposition via difference-of-convex principal component analysis
This page was built for publication: Subgrid modelling for two-dimensional turbulence using neural networks
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4559252)
