Multi-Task Learning based Convolutional Models with Curriculum Learning for the Anisotropic Reynolds Stress Tensor in Turbulent Duct Flow

The Reynolds-averaged Navier-Stokes (RANS) equations require accurate modeling of the anisotropic Reynolds stress tensor. Traditional closure models, while sophisticated, often only apply to restricted flow configurations. Researchers have started using machine learning approaches to tackle this pro...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2022-01
Main Authors: Haitz Sáez de Ocáriz Borde, Sondak, David, Protopapas, Pavlos
Format: Article
Language:English
Subjects:
Artificial neural networks
Computer architecture
Curricula
Fluid dynamics
Machine learning
Mathematical analysis
Modelling
Neural networks
Reynolds averaged Navier-Stokes method
Reynolds stress
Tensors
Turbulence
Turbulent flow
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The Reynolds-averaged Navier-Stokes (RANS) equations require accurate modeling of the anisotropic Reynolds stress tensor. Traditional closure models, while sophisticated, often only apply to restricted flow configurations. Researchers have started using machine learning approaches to tackle this problem by developing more general closure models informed by data. In this work we build upon recent convolutional neural network architectures used for turbulence modeling and propose a multi-task learning-based fully convolutional neural network that is able to accurately predict the normalized anisotropic Reynolds stress tensor for turbulent duct flows. Furthermore, we also explore the application of curriculum learning to data-driven turbulence modeling.
ISSN:2331-8422