Improving CFD simulations by local machine-learned correction

TL;DR

This paper introduces a machine learning-based correction method that improves the accuracy of coarse-mesh CFD simulations, enabling faster computations without sacrificing solution quality, demonstrated on turbulent channel flow.

Contribution

A novel machine learning approach predicts and corrects discretization errors in coarse CFD meshes, enhancing accuracy and computational efficiency.

Findings

Achieved stable 3D turbulent flow simulations with corrected coarse meshes.

Demonstrated significant speedup without loss of accuracy.

Validated method's effectiveness on engineering flow problems.

Abstract

High-fidelity computational fluid dynamics (CFD) simulations for design space explorations can be exceedingly expensive due to the cost associated with resolving the finer scales. This computational cost/accuracy trade-off is a major challenge for modern CFD simulations. In the present study, we propose a method that uses a trained machine learning model that has learned to predict the discretization error as a function of largescale flow features to inversely estimate the degree of lost information due to mesh coarsening. This information is then added back to the low-resolution solution during runtime, thereby enhancing the quality of the under-resolved coarse mesh simulation. The use of a coarser mesh produces a non-linear benefit in speed while the cost of inferring and correcting for the lost information has a linear cost. We demonstrate the numerical stability of a problem of engineering interest, a 3D turbulent channel flow. In addition to this demonstration, we further show the potential for speedup without sacrificing solution accuracy using this method, thereby making the cost/accuracy trade-off of CFD more favorable.