Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

Machine-learned closure of URANS for stably stratified turbulence: connecting physical timescales & data hyperparameters of deep time-series models

Journal Article · · Machine Learning: Science and Technology
 [1];  [2];  [3];  [1];  [1];  [4];  [5]
  1. Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
  2. Univ. of Massachusetts, Amherst, MA (United States); Massachusetts Inst. of Technology (MIT), Lexington, MA (United States). Lincoln Lab.
  3. Univ. of Massachusetts, Amherst, MA (United States); US Dept. of Defense, Vicksburg, MS (United States). High Performance Computing Modernization Program
  4. Univ. of Washington, Seattle, WA (United States)
  5. Univ. of Massachusetts, Amherst, MA (United States)
+ Show Author Affiliations
Stably stratified turbulence (SST), a model that is representative of the turbulence found in the oceans and atmosphere, is strongly affected by fine balances between forces and becomes more anisotropic in time for decaying scenarios. Moreover, there is a limited understanding of the physical phenomena described by some of the terms in the Unsteady Reynolds-Averaged Navier–Stokes (URANS) equations—used to numerically simulate approximate solutions for such turbulent flows. Rather than attempting to model each term in URANS separately, it is attractive to explore the capability of machine learning (ML) to model groups of terms, i.e. to directly model the force balances. We develop deep time-series ML for closure modeling of the URANS equations applied to SST. We consider decaying SST which are homogeneous and stably stratified by a uniform density gradient, enabling dimensionality reduction. We consider two time-series ML models: long short-term memory and neural ordinary differential equation. Both models perform accurately and are numerically stable in a posteriori (online) tests. Furthermore, we explore the data requirements of the time-series ML models by extracting physically relevant timescales of the complex system. We find that the ratio of the timescales of the minimum information required by the ML models to accurately capture the dynamics of the SST corresponds to the Reynolds number of the flow. The current framework provides the backbone to explore the capability of such models to capture the dynamics of high-dimensional complex dynamical system like SST flows.
Research Organization:
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)
Sponsoring Organization:
US Department of the Navy, Office of Naval Research (ONR); USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
Grant/Contract Number:
AC05-00OR22725
OSTI ID:
2483444
Journal Information:
Machine Learning: Science and Technology, Journal Name: Machine Learning: Science and Technology Journal Issue: 4 Vol. 5; ISSN 2632-2153
Publisher:
IOP PublishingCopyright Statement
Country of Publication:
United States
Language:
English

References (59)

Multilayer feedforward networks are universal approximators journal January 1989
Learning subgrid-scale models with neural ordinary differential equations journal July 2023
Recurrent Neural Networks for Time Series Forecasting: Current status and future directions journal January 2021
Temporal Fusion Transformers for interpretable multi-horizon time series forecasting journal October 2021
Turbulence control in plane Couette flow using low-dimensional neural ODE-based models and deep reinforcement learning journal June 2023
Data-driven turbulence modelling of inherently unsteady flow in stratified water storage tanks journal February 2024
Stabilized neural ordinary differential equations for long-time forecasting of dynamical systems journal February 2023
Interpretable spatio-temporal attention LSTM model for flood forecasting journal August 2020
A review of irregular time series data handling with gated recurrent neural networks journal June 2021
A survey on long short-term memory networks for time series prediction journal January 2021
Linear logistic regression with weight thresholding for flow regime classification of a stratified wake journal March 2023
The energy cascade in a strongly stratified fluid journal February 2006
Diffusion in stably stratified turbulence journal December 1996
The effects of stable stratification on the decay of initially isotropic homogeneous turbulence journal December 2018
Deep learning of mixing by two ‘atoms’ of stratified turbulence journal January 2019
Mixing across stable density interfaces in forced stratified turbulence journal April 2023
A data-driven method for modelling dissipation rates in stratified turbulence journal December 2023
Prandtl number effects on extreme mixing events in forced stratified turbulence journal March 2024
Relationship between vertical shear rate and kinetic energy dissipation rate in stably stratified flows journal January 2006
Could Machine Learning Break the Convection Parameterization Deadlock? journal June 2018
Spatially Extended Tests of a Neural Network Parametrization Trained by Coarse‐Graining journal August 2019
Correcting Weather and Climate Models by Machine Learning Nudged Historical Simulations journal August 2021
Data‐Driven Identification of Turbulent Oceanic Mixing From Observational Microstructure Data journal December 2021
Correcting Coarse‐Grid Weather and Climate Models by Machine Learning From Global Storm‐Resolving Simulations journal February 2022
Deep Learning for Subgrid‐Scale Turbulence Modeling in Large‐Eddy Simulations of the Convective Atmospheric Boundary Layer journal May 2022
Deep learning journal May 2015
Stable machine-learning parameterization of subgrid processes for climate modeling at a range of resolutions journal July 2020
Recurrent Neural Networks for Multivariate Time Series with Missing Values journal April 2018
Dynamics of turbulence strongly influenced by buoyancy journal July 2003
Predicting turbulence in flows with strong stable stratification journal January 2006
A unifying framework for parameterizing stably stratified shear-flow turbulence journal March 2014
Wall-modeled large-eddy simulations of spanwise rotating turbulent channels—Comparing a physics-based approach and a data-based approach journal December 2019
Assessment of unsteady flow predictions using hybrid deep learning based reduced-order models journal January 2021
Neural ODE to model and prognose thermoacoustic instability journal January 2022
Data-driven reduced-order modeling of spatiotemporal chaos with neural ordinary differential equations journal July 2022
Generalizability of transformer-based deep learning for multidimensional turbulent flow data journal February 2024
Deep learning to represent subgrid processes in climate models journal September 2018
Phenomenology of two-dimensional stably stratified turbulence under large-scale forcing journal January 2017
The effect of Prandtl number on decaying stratified turbulence journal February 2023
The application of machine learning to predictions of optical turbulence in the surface layer at Baikal Astrophysical Observatory journal May 2021
Data-driven control of spatiotemporal chaos with reduced-order neural ODE-based models and reinforcement learning
  • Zeng, Kevin; Linot, Alec J.; Graham, Michael D.
  • Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 478, Issue 2267 https://doi.org/10.1098/rspa.2022.0297
journal November 2022
Energy spectrum of buoyancy-driven turbulence journal August 2014
Data-driven classification of sheared stratified turbulence from experimental shadowgraphs journal March 2024
Recurrent neural networks and robust time series prediction journal March 1994
Deep Learning with Long Short-Term Memory for Time Series Prediction journal June 2019
Second Moment Closure Modeling and Direct Numerical Simulation of Stratified Shear Layers journal February 2022
A Study of Second Moment Closure Modeling for Stratified Wakes Using DNS Ensembles
  • Jain, Naman; Huang, Xinyi; Yang, Xiang
  • Volume 1: Fluid Applications and Systems (FASTC); Fluid Measurement and Instrumentation (FMITC); Fluid Mechanics (FMTC) https://doi.org/10.1115/FEDSM2022-86979
conference August 2022
Study on the Applicability of a Machine-Learning Framework to Improve Modeling for the Stratification Phenomenon conference August 2022
The pendulum—Rich physics from a simple system journal February 1986
Constructing Neural Network Based Models for Simulating Dynamical Systems journal February 2023
Turbulence Modeling in the Age of Data journal January 2019
Machine Learning for Fluid Mechanics journal September 2019
Fluid Motions in the Presence of Strong Stable Stratification journal January 2000
Using Ensemble of Neural Networks to Learn Stochastic Convection Parameterizations for Climate and Numerical Weather Prediction Models from Data Simulated by a Cloud Resolving Model journal January 2013
Long Short-Term Memory journal November 1997
A Review of Recurrent Neural Networks: LSTM Cells and Network Architectures journal July 2019
Spatially Local Surrogate Modeling of Subgrid-Scale Effects in Idealized Atmospheric Flows: A Deep Learned Approach Using High-Resolution Simulation Data journal October 2024
Unprecedented cloud resolution in a GPU-enabled full-physics atmospheric climate simulation on OLCF’s summit supercomputer journal July 2021
Neural Ordinary Differential Equations for Nonlinear System Identification conference June 2022

Similar Records

Implications of inertial subrange scaling for stably stratified mixing
Journal Article · Wed Mar 23 20:00:00 EDT 2022 · Journal of Fluid Mechanics · OSTI ID:1879371
The myth of URANS
Journal Article · Sun Jun 18 20:00:00 EDT 2023 · Journal of Turbulence (Online) · OSTI ID:1994133
On predicting the transition to turbulence in stably stratified fluids
Conference · Fri Aug 01 00:00:00 EDT 1997 · OSTI ID:519140

Related Subjects

97 MATHEMATICS AND COMPUTING
deep time-series models
stably stratified turbulence
timescale extraction
unsteady RANS