Physics-Informed Neural Networks for Multiphysics Data Assimilation with Application to Subsurface Transport

Abstract

Data assimilation for parameter and state estimation in subsurface transport problems remains a significant challenge due to the sparsity of measurements, the heterogeneity of porous media, and the high computational cost of forward numerical models. We present a physics-informed deep neural networks (DNNs) machine learning method for estimating space-dependent hydraulic conductivity, hydraulic head, and concentration fields from sparse measurements. In this approach, we employ individual DNNs to approximate the unknown parameters (e.g., hydraulic conductivity) and states (e.g., hydraulic head and concentration) of a physical system, and jointly train these DNNs by minimizing the loss function that consists of the governing equations residuals in addition to the error with respect to measurement data. We apply this approach to assimilate conductivity, hydraulic head, and concentration measurements for joint inversion of the conductivity, hydraulic head, and concentration fields in a steady-state advection--dispersion problem. We study the accuracy of the physics-informed DNN approach with respect to data size, number of variables (conductivity and head versus conductivity, head, and concentration), DNNs size, and DNN initialization during training. We demonstrate that the physics-informed DNNs are significantly more accurate than standard data-driven DNNs when the training set consists of sparse data. We also show that the accuracy of parameter estimation increases as additional variables are inverted jointly.

References in corpus (9)

• Adam: A Method for Stochastic Optimization
• Discovering governing equations from data: Sparse identification of nonlinear dynamical systems
• DeepXDE: A deep learning library for solving differential equations
• Artificial Neural Networks for Solving Ordinary and Partial Differential Equations
• Proceedings of the 29th International Conference on Machine Learning (ICML-12)
• A unified deep artificial neural network approach to partial differential equations in complex geometries
• Physics-Driven Regularization of Deep Neural Networks for Enhanced Engineering Design and Analysis
• Learning Parameters and Constitutive Relationships with Physics Informed Deep Neural Networks
• The Neural Network Approach to Inverse Problems in Differential Equations

Cited by in corpus (39)

• DeepXDE: A deep learning library for solving differential equations
• Differentiable modeling to unify machine learning and physical models and advance Geosciences
• Physics-Informed Neural Network for Modelling the Thermochemical Curing Process of Composite-Tool Systems During Manufacture
• A Physics Informed Neural Network for Time-Dependent Nonlinear and Higher Order Partial Differential Equations
• Physics-Informed Neural Network Method for Forward and Backward Advection-Dispersion Equations
• Learning Deep Implicit Fourier Neural Operators (IFNOs) with Applications to Heterogeneous Material Modeling
• Physics-informed neural networks of the Saint-Venant equations for downscaling a large-scale river model
• CCSNet: a deep learning modeling suite for CO${}_2$ storage
• Deep-Learning based Inverse Modeling Approaches: A Subsurface Flow Example
• A Critical Review of Physics-Informed Machine Learning Applications in Subsurface Energy Systems
• A novel meta-learning initialization method for physics-informed neural networks
• Learning Unknown Physics of non-Newtonian Fluids
• Surrogate and inverse modeling for two-phase flow in porous media via theory-guided convolutional neural network
• Pre-training strategy for solving evolution equations based on physics-informed neural networks
• Use of Multifidelity Training Data and Transfer Learning for Efficient Construction of Subsurface Flow Surrogate Models
• Multi-Output Physics-Informed Neural Networks for Forward and Inverse PDE Problems with Uncertainties
• Data vs. Physics: The Apparent Pareto Front of Physics-Informed Neural Networks
• Bayesian Physics Informed Neural Networks for Data Assimilation and Spatio-Temporal Modelling of Wildfires
• Hybrid Machine Learning for Scanning Near-field Optical Spectroscopy
• Physics informed deep learning for computational elastodynamics without labeled data
• Physics-Informed Neural Network Method for Parabolic Differential Equations with Sharply Perturbed Initial Conditions
• Physics-Informed Machine Learning Method for Large-Scale Data Assimilation Problems
• Time Series Predictions in Unmonitored Sites: A Survey of Machine Learning Techniques in Water Resources
• A Mathematical Guide to Operator Learning
• A Deep Learning Approach for the solution of Probability Density Evolution of Stochastic Systems
• A Dimension-Augmented Physics-Informed Neural Network (DaPINN) with High Level Accuracy and Efficiency
• Flow reconstruction by multiresolution optimization of a discrete loss with automatic differentiation
• A mixed formulation for physics-informed neural networks as a potential solver for engineering problems in heterogeneous domains: comparison with finite element method
• Forward and Inverse Simulation of Pseudo-Two-Dimensional Model of Lithium-Ion Batteries Using Neural Networks
• Meshless physics-informed deep learning method for three-dimensional solid mechanics
• Neural optical flow for planar and stereo PIV
• Multi-Constitutive Neural Network for Large Deformation Poromechanics Problem
• Are Two Hidden Layers Still Enough for the Physics-Informed Neural Networks?
• Compatible finite element interpolated neural networks
• On the convergence of PINNs
• About optimal loss function for training physics-informed neural networks under respecting causality
• Applying physics-based loss functions to neural networks for improved generalizability in mechanics problems
• A deep learning energy method for hyperelasticity and viscoelasticity
• Physics-informed neural networks for non-Newtonian fluid thermo-mechanical problems: an application to rubber calendering process