Reconstructing unsteady flows from sparse, noisy measurements with a physics-constrained convolutional neural network

Abstract

Data from fluid flow measurements are typically sparse, noisy, and heterogeneous, often from mixed pressure and velocity measurements, resulting in incomplete datasets. In this paper, we develop a physics-constrained convolutional neural network, which is a deterministic tool, to reconstruct the full flow field from incomplete data. We explore three loss functions, both from machine learning literature and newly proposed: (i) the softly-constrained loss, which allows the prediction to take any value; (ii) the snapshot-enforced loss, which constrains the prediction at the sensor locations; and (iii) the mean-enforced loss, which constrains the mean of the prediction at the sensor locations. The proposed methods do not require the full flow field during training, making it suitable for reconstruction from incomplete data. We apply the method to reconstruct a laminar wake of a bluff body and a turbulent Kolmogorov flow. First, we assume that measurements are not noisy and reconstruct both the laminar wake and the Kolmogorov flow from sensors located at fewer than 1% of all grid points. The snapshot-enforced loss reduces the reconstruction error of the Kolmogorov flow by approximately 25% compared to the softly-constrained loss. Second, we assume that measurements are noisy and propose the mean-enforced loss to reconstruct the laminar wake and the Kolmogorov flow at three different signal-to-noise ratios. We find that, across the ratios tested, the loss functions with harder constraints are more robust to both the random initialization of the networks and the noise levels in the measurements. At high noise levels, the mean-enforced loss can recover the instantaneous snapshots accurately, making it the suitable choice when reconstructing flows from data corrupted with an unknown amount of noise. The proposed method opens opportunities for physical flow reconstruction from sparse, noisy data.

References in corpus (22)

• Super-resolution reconstruction of turbulent flows with machine learning
• Machine learning based spatio-temporal super resolution reconstruction of turbulent flows
• Unsupervised deep learning for super-resolution reconstruction of turbulence
• Uncovering near-wall blood flow from sparse data with physics-informed neural networks
• Shallow Neural Networks for Fluid Flow Reconstruction with Limited Sensors
• Super-resolution and denoising of fluid flow using physics-informed convolutional neural networks without high-resolution labels
• Robust flow field reconstruction from limited measurements via sparse representation
• Super-Resolution Analysis via Machine Learning: A Survey for Fluid Flows
• Least Squares Shadowing sensitivity analysis of chaotic limit cycle oscillations
• Probabilistic neural networks for fluid flow surrogate modeling and data recovery
• Learning in Sinusoidal Spaces with Physics-Informed Neural Networks
• Reconstruction of turbulent data with deep generative models for semantic inpainting from TURB-Rot database
• State estimation in turbulent channel flow from limited observations
• Mean-flow Data Assimilation based on minimal correction of turbulence models: application to turbulent high-Reynolds number backward-facing step
• Gappy spectral proper orthogonal decomposition
• Inferring unknown unknowns: Regularized bias-aware ensemble Kalman filter
• Multi-scale data reconstruction of turbulent rotating flows with Gappy POD, Extended POD and Generative Adversarial Networks
• ModalPINN: an extension of Physics-Informed Neural Networks with enforced truncated Fourier decomposition for periodic flow reconstruction using a limited number of imperfect sensors
• Stability, sensitivity and optimisation of chaotic acoustic oscillations
• Deep learning fluid flow reconstruction around arbitrary two-dimensional objects from sparse sensors using conformal mappings
• Reconstructing three-dimensional bluff body wake from sectional flow fields with convolutional neural networks
• Stochastic Latent Transformer: Efficient Modelling of Stochastically Forced Zonal Jets

Cited by in corpus (2)

• Reconstruction of three-dimensional turbulent flows from sparse and noisy planar measurements: A weight-sharing neural network approach
• Leveraging Scale Separation and Stochastic Closure for Data-Driven Prediction of Chaotic Dynamics