A twin-decoder structure for incompressible laminar flow reconstruction with uncertainty estimation around 2D obstacles

TL;DR

This paper introduces a twin auto-encoder network for reconstructing incompressible laminar flows around 2D obstacles, providing uncertainty estimates to improve robustness and reliability outside the training data domain.

Contribution

A novel twin auto-encoder architecture is proposed for flow reconstruction with integrated uncertainty estimation, enhancing model trustworthiness in CFD surrogate modeling.

Findings

Strong correlation between reconstruction score and prediction error.

Uncertainty estimation enables rejection or confidence intervals for predictions.

Model demonstrates robustness on unseen shapes.

Abstract

Over the past few years, deep learning methods have proved to be of great interest for the computational fluid dynamics community, especially when used as surrogate models, either for flow reconstruction, turbulence modeling, or for the prediction of aerodynamic coefficients. Overall, exceptional levels of accuracy have been obtained, but the robustness and reliability of the proposed approaches remain to be explored, particularly outside of the confidence region defined by the training dataset. In this contribution, we present a twin auto-encoder network for incompressible laminar flow reconstruction with uncertainty estimation around 2D obstacles. The proposed architecture is trained over a dataset composed of numerically-computed laminar flows around 12,000 random shapes, and naturally enforces a quasi-linear relation between a geometric reconstruction branch and the flow prediction decoder. Based on this feature, two uncertainty estimation processes are proposed, allowing either a binary decision (accept or reject prediction), or proposing a confidence interval along with the flow quantities prediction (u,v,p). Results over dataset samples as well as unseen shapes show a strong positive correlation of this reconstruction score to the mean-squared error of the flow prediction. Such approaches offer the possibility to warn the user of trained models when the provided input shows too large deviation from the training data, making the produced surrogate model conservative for fast and reliable flow prediction.