A perspective on machine learning methods in turbulence modeling

A Beck, M Kurz - GAMM‐Mitteilungen, 2021 - Wiley Online Library
This work presents a review of the current state of research in data‐driven turbulence
closure modeling. It offers a perspective on the challenges and open issues but also on the …

Toward neural-network-based large eddy simulation: Application to turbulent channel flow

J Park, H Choi - Journal of Fluid Mechanics, 2021 - cambridge.org
A fully connected neural network (NN) is used to develop a subgrid-scale (SGS) model
mapping the relation between the SGS stresses and filtered flow variables in a turbulent …

A priori analysis on deep learning of subgrid-scale parameterizations for Kraichnan turbulence

S Pawar, O San, A Rasheed, P Vedula - Theoretical and Computational …, 2020 - Springer
In the present study, we investigate different data-driven parameterizations for large eddy
simulation of two-dimensional turbulence in the a priori settings. These models utilize …

Frame invariant neural network closures for Kraichnan turbulence

S Pawar, O San, A Rasheed, P Vedula - Physica A: Statistical Mechanics …, 2023 - Elsevier
Numerical simulations of geophysical and atmospheric flows have to rely on
parameterizations of subgrid scale processes due to their limited spatial resolution. Despite …

Data assimilation empowered neural network parametrizations for subgrid processes in geophysical flows

S Pawar, O San - Physical Review Fluids, 2021 - APS
In the past couple of years, there has been a proliferation in the use of machine learning
approaches to represent subgrid-scale processes in geophysical flows with an aim to …

Large eddy simulation of flow over a circular cylinder with a neural-network-based subgrid-scale model

M Kim, J Park, H Choi - Journal of Fluid Mechanics, 2024 - cambridge.org
A neural-network-based large eddy simulation is performed for flow over a circular cylinder.
To predict the subgrid-scale (SGS) stresses, we train two fully connected neural network …

Modélisation du tenseur de contraintes sous-mailles par réseau de neurones à convolutions 3D en Turbulence homogène isotrope

N Saura - 2021 - theses.hal.science
La modélisation numérique de la turbulence est une des approches classiques pour étudier
la complexité de la dynamique des échelles actives dans un écoulement turbulent. La …

Predicting unresolved scales interactions with 3D neural networks in homogeneous isotropic turbulence

N Saura, T Gomez - Europhysics Letters, 2023 - iopscience.iop.org
Interactions between the different scales of motion featured by any turbulent flow are
currently mathematically and numerically intractable. Instead, numerical reduced models …

Subgrid stress tensor prediction in homogeneous isotropic turbulence using 3D-convolutional neural networks

N Saura, T Gomez - Available at SSRN 4184202 - papers.ssrn.com
Interactions between the different scales of motion featured by any turbulent flow are
currently mathematically and numerically intractable. Instead, numerical reduced models …

Nonlinear System Discovery and Machine Learning for Dynamical Systems

SAS Romeo - 2023 - search.proquest.com
Extracting physics from data has become a crucial task in fields where abundant data is
available. However, the underlying governing equations, physical laws, or models based on …