Physics-Informed Tensor Basis Neural Network for Turbulence Closure Modeling

TL;DR

This paper introduces a physics-informed neural network model for turbulence closure that improves anisotropy tensor predictions in RANS simulations, especially in complex flow scenarios, but does not necessarily enhance mean velocity or pressure predictions.

Contribution

It presents a novel deep learning approach incorporating turbulence physics constraints to improve anisotropy tensor modeling in turbulence simulations.

Findings

Significant improvement in anisotropy tensor predictions over traditional models.

Better performance in complex flows with surface curvature and flow separation.

No consistent improvement in mean velocity and pressure field predictions.

Abstract

Despite the increasing availability of high-performance computational resources, Reynolds-Averaged Navier-Stokes (RANS) simulations remain the workhorse for the analysis of turbulent flows in real-world applications. Linear eddy viscosity models (LEVM), the most commonly employed model type, cannot accurately predict complex states of turbulence. This work combines a deep-neural-network-based, nonlinear eddy viscosity model with turbulence realizability constraints as an inductive bias in order to yield improved predictions of the anisotropy tensor. Using visualizations based on the barycentric map, we show that the proposed machine learning method's anisotropy tensor predictions offer a significant improvement over all LEVMs in traditionally challenging cases with surface curvature and flow separation. However, this improved anisotropy tensor does not, in general, yield improved mean-velocity and pressure field predictions in comparison with the best-performing LEVM.