Deep learning in fluid dynamics

It was only a matter of time before deep neural networks (DNNs) – deep learning – made their mark in turbulence modelling, or more broadly, in the general area of high-dimensional, complex dynamical systems. In the last decade, DNNs have become a dominant data mining tool for big data applications....

Full description

Saved in:
Bibliographic Details
Published in:Journal of fluid mechanics Vol. 814; pp. 1 - 4
Main Author: Kutz, J. Nathan
Format: Journal Article
Language:English
Published: Cambridge, UK Cambridge University Press 10.03.2017
Subjects:
Algorithms
Anisotropy
Applied mathematics
Artificial intelligence
Artificial neural networks
Computational fluid dynamics
Data management
Data mining
Data processing
Datasets
Decomposition
Deep learning
Dynamical systems
Dynamics
Fluid dynamics
Fluid flow
Focus on Fluids
Hydrodynamics
Modelling
Navier-Stokes equations
Neural networks
Physics
Reynolds averaged Navier-Stokes method
Studies
Success
Turbulence
Turbulence models
computational methods
turbulence modelling
low-dimensional models
ISSN:0022-1120, 1469-7645
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:It was only a matter of time before deep neural networks (DNNs) – deep learning – made their mark in turbulence modelling, or more broadly, in the general area of high-dimensional, complex dynamical systems. In the last decade, DNNs have become a dominant data mining tool for big data applications. Although neural networks have been applied previously to complex fluid flows, the article featured here (Ling et al., J. Fluid Mech., vol. 807, 2016, pp. 155–166) is the first to apply a true DNN architecture, specifically to Reynolds averaged Navier Stokes turbulence models. As one often expects with modern DNNs, performance gains are achieved over competing state-of-the-art methods, suggesting that DNNs may play a critically enabling role in the future of modelling complex flows.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0022-1120
1469-7645
DOI:10.1017/jfm.2016.803