Prediction of flow and elastic stresses in a viscoelastic turbulent channel flow using convolutional neural networks

Arivazhagan G. Balasubramanian,Ricardo Vinuesa,Outi Tammisola

2024-08-17

Abstract:Neural-network models have been employed to predict the instantaneous flow close to the wall in a viscoelastic turbulent channel flow. Numerical simulation data at the wall is utilized to predict the instantaneous velocity-fluctuations and polymeric-stress-fluctuations at three different wall-normal positions in the buffer region. The ability of non-intrusive predictions has not been previously investigated in non-Newtonian turbulence. Our analysis shows that velocity-fluctuations are predicted well from wall measurements in viscoelastic turbulence. The models exhibit enhanced accuracy in predicting quantities of interest during the hibernation intervals, facilitating a deeper understanding of the underlying physics during low-drag events. The neural-network models also demonstrate a reasonably good accuracy in predicting polymeric-shear stress and the trace of the polymer stress at a given wall-normal location. This method could be used in flow control or when only wall information is available from experiments (for example, in opaque fluids). More importantly, only velocity and pressure information can be measured experimentally, while polymeric elongation and orientation cannot be directly measured despite their importance for turbulent dynamics. We therefore study the possibility to reconstruct the polymeric-stress fields from velocity or pressure measurements in viscoelastic turbulent flows. The results are promising but also underline that a lack of small scales in the input velocity fields can alter the rate of energy transfer from flow to polymers, affecting the prediction of the polymer-stress fluctuations. The present approach not only aids in extracting polymeric-stress information but also gives information about the link between polymeric-stress and velocity fields in viscoelastic turbulence.

Fluid Dynamics,Machine Learning

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to predict the near - wall flow velocity and elastic stress in viscoelastic turbulent channel flow using convolutional neural networks (CNN). Specifically, the researchers use numerical simulation data to predict the instantaneous velocity fluctuations and polymer stress fluctuations at three different wall - normal positions in the buffer zone. This is the first study on the non - invasive prediction ability in non - Newtonian turbulence. The main objectives of the paper include: 1. **Verify the prediction ability**: Evaluate whether the neural network model can accurately predict the velocity fluctuations and polymer stress fluctuations in viscoelastic turbulence from wall - measurement data. 2. **Analysis of low - drag events**: Study the prediction performance of the model during low - drag events (hibernation intervals), which are of great significance for understanding the underlying physical mechanisms. 3. **Potential for experimental applications**: Explore how to use only wall - information (for example, in opaque fluids) for flow - field control or prediction. 4. **Reconstruct the polymer stress field**: Study the possibility of reconstructing the polymer stress field from velocity or pressure measurements, because the elongation and orientation of polymers are very important in turbulent dynamics but cannot be directly measured. Through these objectives, the paper aims to provide new methods for experimental research and deepen the understanding of the interaction between polymers and the flow field in viscoelastic turbulence.

Prediction of flow and elastic stresses in a viscoelastic turbulent channel flow using convolutional neural networks

Predicting the wall-shear stress and wall pressure through convolutional neural networks

Predicting the near-wall region of turbulence through convolutional neural networks

Uncovering wall-shear stress dynamics from neural-network enhanced fluid flow measurements

A scalable convolutional neural network approach to fluid flow prediction in complex environments

ViscoelasticNet: A physics informed neural network framework for stress discovery and model selection

Prediction of Aerodynamic Flow Fields Using Convolutional Neural Networks

Prediction of turbulent heat transfer using convolutional neural networks

Reconstructing turbulent velocity information for arbitrarily gappy flow fields using the deep convolutional neural network

Machine Learning for the Flow Prediction of Fluids with Memory Effects on the Stress

Modelling pressure-Hessian from local velocity gradients information in an incompressible turbulent flow field using deep neural networks

A Flow Feature Detection Method for Modeling Pressure Distribution Around a Cylinder in Non-Uniform Flows by Using a Convolutional Neural Network

Application of boundary-fitted convolutional neural network to simulate non-Newtonian fluid flow behavior in eccentric annulus

Go with the flow: deep learning methods for autonomous viscosity estimations

Finite volume method network for acceleration of unsteady computational fluid dynamics: non-reacting and reacting flows

Data Augmented Prediction of Reynolds Stresses for Flows Around an Axisymmetric Body of Revolution

Revisiting Tensor Basis Neural Network for Reynolds stress modeling: Application to plane channel and square duct flows

Alternative Artificial Neural Network Structures for Turbulent Flow Velocity Field Prediction

Machine learning for rapid discovery of laminar flow channel wall modifications that enhance heat transfer

Predictions of turbulent shear flows using deep neural networks