Subsurface Flow

Subsurface flow modeling aims to predict fluid movement through porous media, crucial for applications like oil extraction and CO2 storage. Current research heavily utilizes deep learning, employing architectures like convolutional neural networks (CNNs), recurrent neural networks (RNNs), graph neural networks (GNNs), and U-Nets to create surrogate models that significantly accelerate simulations compared to traditional methods. These models are often enhanced by incorporating physics-based constraints to improve accuracy and generalizability, and are increasingly applied to uncertainty quantification and optimization problems, such as optimal well placement. This accelerates decision-making in resource management and environmental remediation.

Papers

November 1, 2024

An Efficient Hierarchical Preconditioner-Learner Architecture for Reconstructing Multi-scale Basis Functions of High-dimensional Subsurface Fluid Flow
Peiqi Li, Jie Chen
Hierarchical Framework Fluid Flow Multi Scale Sparse Neural Preconditioners Subsurface Flow Subsurface Property Subsurface Model

October 26, 2024

GeoFUSE: A High-Efficiency Surrogate Model for Seawater Intrusion Prediction and Uncertainty Reduction
Su Jiang, Chuyang Liu, Dipankar Dwivedi
Human Prediction Surrogate Modeling Groundwater Flow Subsurface Flow

August 20, 2024

Accelerated training of deep learning surrogate models for surface displacement and flow, with application to MCMC-based history matching of CO2 storage operations
Yifu Han, Francois P. Hamon, Louis J. Durlofsky
Surrogate Model Surrogate Modeling Faster Training Geological Carbon Storage Subsurface Flow

December 14, 2023

Graph Network Surrogate Model for Subsurface Flow Optimization
Haoyu Tang, Louis J. Durlofsky
Oil Well Geological Carbon Storage Reservoir System Permeability Field Subsurface Flow

June 16, 2023

Learning CO$_2$ plume migration in faulted reservoirs with Graph Neural Networks
Xin Ju, François P. Hamon, Gege Wen, Rayan Kanfar, Mauricio Araya-Polo, Hamdi A. Tchelepi
Graph Neural Network Reservoir Dynamic Gas Plume Graph Based Neural Reservoir System Subsurface Flow

May 15, 2023

Physics-informed Convolutional Recurrent Surrogate Model for Reservoir Simulation with Well Controls
Jungang Chen, Eduardo Gildin, John E. Killough
Surrogate Model Reservoir Dynamic Reservoir Simulation Well Control Subsurface Flow

October 5, 2022

Advanced Deep Learning Architectures for Accurate Detection of Subsurface Tile Drainage Pipes from Remote Sensing Images
Tom-Lukas Breitkopf, Leonard W. Hackel, Mahdyar Ravanbakhsh, Anne-Karin Cooke, Sandra Willkommen, Stefan Broda, Begüm Demir
Remote Sensing Image Deep Learning Architecture Accurate Detection Land Use Subsurface Flow

June 15, 2022

Learning Large-scale Subsurface Simulations with a Hybrid Graph Network Simulator
Tailin Wu, Qinchen Wang, Yinan Zhang, Rex Ying, Kaidi Cao, Rok Sosič, Ridwan Jalali, Hassan Hamam, Marko Maucec, Jure Leskovec
Reservoir Pressure Graph Network Simulator Reservoir Simulation Subsurface Flow Subsurface Model

May 28, 2022

Uncertainty quantification of two-phase flow in porous media via coupled-TgNN surrogate model
Jian Li, Dongxiao Zhang, Tianhao He, Qiang Zheng
Uncertainty Quantification Surrogate Model Multiphase Flow Data Driven Surrogate Porous Medium Subsurface Flow

March 24, 2022

Deep reinforcement learning for optimal well control in subsurface systems with uncertain geology
Yusuf Nasir, Louis J. Durlofsky
Deep Reinforcement Learning Well Control Subsurface Flow Continuous Control Policy

November 14, 2021

Uncertainty quantification and inverse modeling for subsurface flow in 3D heterogeneous formations using a theory-guided convolutional encoder-decoder network
Rui Xu, Dongxiao Zhang, Nanzhe Wang
Uncertainty Quantification Surrogate Model Inverse Modeling Permeability Field Subsurface Flow

Subsurface Flow

Papers

An Efficient Hierarchical Preconditioner-Learner Architecture for Reconstructing Multi-scale Basis Functions of High-dimensional Subsurface Fluid Flow

GeoFUSE: A High-Efficiency Surrogate Model for Seawater Intrusion Prediction and Uncertainty Reduction

Accelerated training of deep learning surrogate models for surface displacement and flow, with application to MCMC-based history matching of CO2 storage operations

Graph Network Surrogate Model for Subsurface Flow Optimization

Learning CO$_2$ plume migration in faulted reservoirs with Graph Neural Networks

Physics-informed Convolutional Recurrent Surrogate Model for Reservoir Simulation with Well Controls

Advanced Deep Learning Architectures for Accurate Detection of Subsurface Tile Drainage Pipes from Remote Sensing Images

Learning Large-scale Subsurface Simulations with a Hybrid Graph Network Simulator

Uncertainty quantification of two-phase flow in porous media via coupled-TgNN surrogate model

Deep reinforcement learning for optimal well control in subsurface systems with uncertain geology

Uncertainty quantification and inverse modeling for subsurface flow in 3D heterogeneous formations using a theory-guided convolutional encoder-decoder network