A Physics-informed Diffusion Model for High-fidelity Flow Field Reconstruction

TL;DR

This paper introduces a physics-informed diffusion model that reconstructs high-fidelity fluid flow data using only high-fidelity training data, capable of handling various input types and leveraging PDE information for improved accuracy.

Contribution

The proposed model uniquely reconstructs high-fidelity flow fields without relying on low-fidelity training data, incorporating physics-informed conditioning for enhanced precision.

Findings

Accurately reconstructs 2D turbulent flows from various input sources.

Operates effectively without retraining for different input types.

Utilizes PDE information to improve reconstruction accuracy.

Abstract

Machine learning models are gaining increasing popularity in the domain of fluid dynamics for their potential to accelerate the production of high-fidelity computational fluid dynamics data. However, many recently proposed machine learning models for high-fidelity data reconstruction require low-fidelity data for model training. Such requirement restrains the application performance of these models, since their data reconstruction accuracy would drop significantly if the low-fidelity input data used in model test has a large deviation from the training data. To overcome this restraint, we propose a diffusion model which only uses high-fidelity data at training. With different configurations, our model is able to reconstruct high-fidelity data from either a regular low-fidelity sample or a sparsely measured sample, and is also able to gain an accuracy increase by using physics-informed conditioning information from a known partial differential equation when that is available. Experimental results demonstrate that our model can produce accurate reconstruction results for 2d turbulent flows based on different input sources without retraining.