Reconstructing High-resolution Turbulent Flows Using Physics-Guided Neural Networks

TL;DR

This paper introduces a physics-guided neural network approach that enhances the reconstruction of high-resolution turbulent flow data from lower-resolution simulations, effectively capturing fine flow details and dynamics.

Contribution

The authors develop a novel super-resolution method incorporating physical relationships and hierarchical generative models to improve DNS data reconstruction from LES predictions.

Findings

Outperforms existing methods in pixel-wise and structural similarity metrics.

Successfully captures fine-scale flow dynamics in turbulent flows.

Effective in both single-snapshot and cross-time experiments.

Abstract

Direct numerical simulation (DNS) of turbulent flows is computationally expensive and cannot be applied to flows with large Reynolds numbers. Large eddy simulation (LES) is an alternative that is computationally less demanding, but is unable to capture all of the scales of turbulent transport accurately. Our goal in this work is to build a new data-driven methodology based on super-resolution techniques to reconstruct DNS data from LES predictions. We leverage the underlying physical relationships to regularize the relationships amongst different physical variables. We also introduce a hierarchical generative process and a reverse degradation process to fully explore the correspondence between DNS and LES data. We demonstrate the effectiveness of our method through a single-snapshot experiment and a cross-time experiment. The results confirm that our method can better reconstruct high-resolution DNS data over space and over time in terms of pixel-wise reconstruction error and structural similarity. Visual comparisons show that our method performs much better in capturing fine-level flow dynamics.