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

Journal Article · 12 December 2024 · Machine Learning: Science and Technology

DOI: https://doi.org/10.1088/2632-2153/ad9704 · OSTI ID:2483444

Gopalakrishnan Meena, Muralikrishnan  ^[1]; Liousas, Demetri  ^[2]; Simin, Andrew D.  ^[3]; Kashi, Aditya  ^[1]; Brewer, Wesley H.  ^[1]; Riley, James J.  ^[4]; de Bruyn Kops, Stephen M.  ^[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)

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