Volume 143, Number 2, July 2023
|Number of page(s)||7|
|Section||Fluid and nonlinear dynamics|
|Published online||24 July 2023|
Predicting unresolved scales interactions with 3D neural networks in homogeneous isotropic turbulence
1 Aix-Marseille Université, CNRS, PIIM, UMR 7345 - Marseille, France
2 Univ. Lille, CNRS, ONERA, Arts et Metiers ParisTech, Centrale Lille, FRE 2017- LMFL - Laboratoire de Mécanigue des Fluids de Lille-Kampé de Fériet - F-59000 Lille, France
3 ICAM Engineering School, Grand Paris Sud Campus - Lieusaint, France
4 CNRS, Insitut Jean le Rond d'Alembert, Sorbonne Université - Paris, France
(a) E-mail: email@example.com (corresponding author)
Received: 3 February 2023
Accepted: 7 June 2023
Interactions between the different scales of motion featured by any turbulent flow are currently mathematically and numerically intractable. Instead, numerical reduced models, such as Large Eddy Simulations (LES), have been proposed: large-scale motions are resolved using the large eddy equations whereas small-scale influence is modeled through the subgrid stress tensor and injected into the large-scale dynamics. In this paper, we develop the learning of this tensor from the raw 3D filtered velocity field diced into sub-cubes whose length is turbulence-induced. We used the U-net convolutional neural network architecture. The performance is assessed using component-wise correlations, PDF and contours comparisons. We extended our a priori analyses to monitor the impacts of such predictions on momentum and kinetic energy evolution. Our model is shown to perform well considering velocity fields extracted from 150% more turbulent simulations.
© 2023 EPLA
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.