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

Direct numerical simulation (DNS) of turbulent flows is compu- tationally 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 degra- dation 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 experi- ment. The results confirm that our method can better reconstruct high-resolution DNS data over space and over time.