Super-resolution and denoising of fluid flow using physics-informed convolutional neural networks without high-resolution labels

Abstract

High-resolution (HR) information of fluid flows, although preferable, is usually less accessible due to limited computational or experimental resources. In many cases, fluid data are generally sparse, incomplete, and possibly noisy. How to enhance spatial resolution and decrease the noise level of flow data is essential and practically useful. Deep learning (DL) techniques have been demonstrated to be effective for super-resolution (SR) tasks, which, however, primarily rely on sufficient HR labels for training. In this work, we present a novel physics-informed DL-based SR solution using convolutional neural networks (CNN), which is able to produce HR flow fields from low-resolution (LR) inputs in high-dimensional parameter space. By leveraging the conservation laws and boundary conditions of fluid flows, the CNN-SR model is trained without any HR labels. Moreover, the proposed CNN-SR solution unifies the forward SR and inverse data assimilation for the scenarios where the physics is partially known, e.g., unknown boundary conditions. Several flow SR problems relevant to cardiovascular applications have been studied to demonstrate the proposed method's effectiveness and merit. Both Gaussian and non-Gaussian MRI noises are investigated to illustrate the denoising capability.

arXiv comment34 pages, 11 figures

References in corpus (23)

• Machine Learning for Fluid Mechanics
• On Dynamic Mode Decomposition: Theory and Applications
• Turbulence Modeling in the Age of Data
• Surrogate Modeling for Fluid Flows Based on Physics-Constrained Deep Learning Without Simulation Data
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• PhyGeoNet: Physics-Informed Geometry-Adaptive Convolutional Neural Networks for Solving Parameterized Steady-State PDEs on Irregular Domain
• A Physics Informed Machine Learning Approach for Reconstructing Reynolds Stress Modeling Discrepancies Based on DNS Data
• PDE-Net 2.0: Learning PDEs from Data with A Numeric-Symbolic Hybrid Deep Network
• Data-Driven Sparse Sensor Placement for Reconstruction
• Modeling the Dynamics of PDE Systems with Physics-Constrained Deep Auto-Regressive Networks
• Quantifying and Reducing Model-Form Uncertainties in Reynolds-Averaged Navier-Stokes Simulations: A Data-Driven, Physics-Based Bayesian Approach
• Robust flow field reconstruction from limited measurements via sparse representation
• Probabilistic neural networks for fluid flow surrogate modeling and data recovery
• Synthetic turbulent inflow generator using machine learning
• Distributed physics informed neural network for data-efficient solution to partial differential equations
• Deep unsupervised learning of turbulence for inflow generation at various Reynolds numbers
• Prediction of Reynolds Stresses in High-Mach-Number Turbulent Boundary Layers using Physics-Informed Machine Learning
• A Multi-Pass GAN for Fluid Flow Super-Resolution
• Data-Driven CFD Modeling of Turbulent Flows Through Complex Structures
• Non-intrusive model reduction of large-scale, nonlinear dynamical systems using deep learning
• Multi-fidelity Generative Deep Learning Turbulent Flows
• A Bi-fidelity Ensemble Kalman Method for PDE-Constrained Inverse Problems
• A particle filter to reconstruct a free-surface flow from a depth camera

Cited by in corpus (15)

• Physics-informed graph neural Galerkin networks: A unified framework for solving PDE-governed forward and inverse problems
• Uncovering near-wall blood flow from sparse data with physics-informed neural networks
• Super-Resolution Analysis via Machine Learning: A Survey for Fluid Flows
• Physics-informed PointNet: A deep learning solver for steady-state incompressible flows and thermal fields on multiple sets of irregular geometries
• Experimental velocity data estimation for imperfect particle images using machine learning
• Multi-resolution partial differential equations preserved learning framework for spatiotemporal dynamics
• Physics-informed data based neural networks for two-dimensional turbulence
• Reconstructing unsteady flows from sparse, noisy measurements with a physics-constrained convolutional neural network
• Generating extreme quantum scattering in graphene with machine learning
• Training Deep Physics-Informed Kolmogorov-Arnold Networks
• Super-resolution reconstruction of turbulent flows from a single Lagrangian trajectory
• Flow field tomography with uncertainty quantification using a Bayesian physics-informed neural network
• Physics-informed Dyna-Style Model-Based Deep Reinforcement Learning for Dynamic Control
• Reconstruction of three-dimensional turbulent flows from sparse and noisy planar measurements: A weight-sharing neural network approach
• Guided Unconditional and Conditional Generative Models for Super-Resolution and Inference of Quasi-Geostrophic Turbulence