Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Physics-Informed Neural Network Method for Forward and Backward Advection-Dispersion Equations

Journal Article · · Water Resources Research
DOI:https://doi.org/10.1029/2020WR029479· OSTI ID:1811960

Advection-dispersion equations (ADEs) are commonly used to describe transport phenomena in porous media. Even though mature discretization-based numerical methods for ADEs exist, some challenges still remain, especially when it comes to solving advection-dominated forward ADEs and diffusion-dominated backward ADEs. The latter problem usually arises in the source identification context and leads to numerically unstable grid-based solutions that require a form of regularization or should be treated as an inverse problem that is computationally more expensive because it requires solving the forward problem multiple times. In this study, we propose a discretization-free approach based on the physics-informed neural network (PINN) method for solving coupled ADE and Darcy flow equations with space-dependent hydraulic conductivity. In this approach, the hydraulic conductivity, hydraulic head, and concentration fields are approximated with deep neural networks (DNNs). We assume that the conductivity field is given by its values on a grid, and we use these values to train the conductivity DNN. The head and concentration DNNs are trained by minimizing the residuals of the flow equation and ADE and using the initial and boundary conditions as additional constraints. The PINN method is applied to one- and two-dimensional forward ADE problems, where its performance for various P\'{e}clet numbers ($Pe$$) is compared with the analytical and numerical solutions. We find that the PINN method is accurate with errors of less than 1\% and outperforms some conventional discretization-based methods for $$Pe$ larger that 100. Next, we demonstrate that the PINN method remains accurate for the backward ADEs, with the relative errors in most cases staying under 5\% compared to the reference concentration field. Finally, we show that when available, the concentration measurements can be easily incorporated in the PINN method and significantly improve (by more than 50\% in the considered cases) the accuracy of the PINN solution of the backward ADE.

Research Organization:
Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
Sponsoring Organization:
USDOE
DOE Contract Number:
AC05-76RL01830
OSTI ID:
1811960
Report Number(s):
PNNL-SA-158769
Journal Information:
Water Resources Research, Vol. 57, Issue 7
Country of Publication:
United States
Language:
English

References (49)

A stable Petrov-Galerkin method for convection-dominated problems January 1997
State of the Art Report on Mathematical Methods for Groundwater Pollution Source Identification January 2001
A unified deep artificial neural network approach to partial differential equations in complex geometries November 2018
A high-order discontinuous Galerkin method for unsteady advection–diffusion problems March 2017
Streamline upwind/Petrov-Galerkin formulations for convection dominated flows with particular emphasis on the incompressible Navier-Stokes equations September 1982
A Limited Memory Algorithm for Bound Constrained Optimization September 1995
Flow over an espresso cup: inferring 3-D velocity and pressure fields from tomographic background oriented Schlieren via physics-informed neural networks March 2021
High order finite difference numerical methods for time-dependent convection–dominated problems November 2005
Approximation by superpositions of a sigmoidal function December 1989
A new difference scheme with high accuracy and absolute stability for solving convection–diffusion equations August 2009
Physics Informed Extreme Learning Machine (PIELM)–A rapid method for the numerical solution of partial differential equations May 2020
A summary of numerical methods for time-dependent advection-dominated partial differential equations March 2001
Bubble functions prompt unusual stabilized finite element methods June 1995
Stabilized finite element methods: I. Application to the advective-diffusive model March 1992
An ‘upwind’ finite element scheme for two-dimensional convective transport equation January 1977
Physics-informed neural networks for multiphysics data assimilation with application to subsurface transport July 2020
A multidimensional streamline-based method to simulate reactive solute transport in heterogeneous porous media July 2010
Positive Solution of Two-Dimensional Solute Transport in Heterogeneous Aquifers November 2006
Nodally integrated implicit gradient reproducing kernel particle method for convection dominated problems February 2016
Multilayer feedforward networks are universal approximators January 1989
A finite element solution for the fractional advection–dispersion equation December 2008
A new finite element formulation for computational fluid dynamics: VIII. The galerkin/least-squares method for advective-diffusive equations May 1989
A new finite element formulation for computational fluid dynamics: II. Beyond SUPG March 1986
Conservation properties for the Galerkin and stabilised forms of the advection–diffusion and incompressible Navier–Stokes equations March 2005
Machine learning in cardiovascular flows modeling: Predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networks January 2020
Neural-network methods for boundary value problems with irregular boundaries January 2000
Neural algorithm for solving differential equations November 1990
Monotone finite volume schemes for diffusion equations on unstructured triangular and shape-regular polygonal meshes November 2007
Physics-informed neural networks for high-speed flows March 2020
One-dimensional linear advection–diffusion equation: Analytical and finite element solutions January 2015
Adjoint method for obtaining backward-in-time location and travel time probabilities of a conservative groundwater contaminant November 1999
An upstream finite element method for solution of transient transport equation in fractured porous media June 1982
A third-order semi-implicit finite difference method for solving the one-dimensional convection-diffusion equation July 1988
fPINNs: Fractional Physics-Informed Neural Networks January 2019
Contaminant transport through porous media: An overview of experimental and numerical studies March 2014
Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations February 2019
High Order Difference Schemes for Unsteady One-Dimensional Diffusion-Convection Problems September 1994
An energy approach to the solution of partial differential equations in computational mechanics via machine learning: Concepts, implementation and applications April 2020
On the Convergence of Physics Informed Neural Networks for Linear Second-Order Elliptic and Parabolic Type PDEs June 2020
DGM: A deep learning algorithm for solving partial differential equations December 2018
Surrogate modeling for fluid flows based on physics-constrained deep learning without simulation data April 2020
Investigating the Effects of Anisotropic Mass Transport on Dendrite Growth in High Energy Density Lithium Batteries November 2015
Physics‐Informed Deep Neural Networks for Learning Parameters and Constitutive Relationships in Subsurface Flow Problems May 2020
Optimal weighting in the finite difference solution of the convection-dispersion equation December 1997
The Deep Ritz Method: A Deep Learning-Based Numerical Algorithm for Solving Variational Problems February 2018
Modeling fluid flow and transport in variably saturated porous media with the STOMP simulator. 1. Nonvolatile three-phase model description January 1995
Two numerical methods for solving a backward heat conduction problem August 2006
Numerical solution of unsteady advection dispersion equation arising in contaminant transport through porous media using neural networks August 2016
Machine learning for metal additive manufacturing: predicting temperature and melt pool fluid dynamics using physics-informed neural networks January 2021

Related Subjects