Multi-Fidelity Data Assimilation For Physics Inspired Machine Learning In Uncertainty Quantification Of Fluid Turbulence Simulations

TL;DR

This paper introduces a physics-inspired machine learning framework using CNNs to predict and optimize perturbations in turbulence models, enhancing uncertainty quantification in fluid simulations.

Contribution

It develops a novel ML-based perturbation method that incorporates physics biases to improve turbulence model error estimates in CFD simulations.

Findings

CNN-based perturbation predictions improve error estimates

The approach modulates perturbation intensity and location effectively

Enhanced uncertainty quantification in turbulence modeling

Abstract

Reliable prediction of turbulent flows is an important necessity across different fields of science and engineering. In Computational Fluid Dynamics (CFD) simulations, the most common type of models are eddy viscosity models that are computationally inexpensive but introduce a high degree of epistemic error. The Eigenspace Perturbation Method (EPM) attempts to quantify this predictive uncertainty via physics based perturbation in the spectral representation of the predictions. While the EPM outlines how to perturb, it does not address how much or even where to perturb. We address this need by introducing machine learning models to predict the details of the perturbation, thus creating a physics inspired machine learning (ML) based perturbation framework. In our choice of ML models, we focus on incorporating physics based inductive biases while retaining computational economy. Specifically we use a Convolutional Neural Network (CNN) to learn the correction between turbulence model predictions and true results. This physics inspired machine learning based perturbation approach is able to modulate the intensity and location of perturbations and leads to improved estimates of turbulence model errors.