U-FNO -- An enhanced Fourier neural operator-based deep-learning model for multiphase flow

2021/09/03 by Gege Wen, Zongyi Li, Wen, Gege +7 · 46 citations
Engineering · Environmental Science · #CO2 Sequestration and Geologic Interactions #Enhanced Oil Recovery Techniques #FOS: Computer and information sciences #FOS: Physical sciences #Geophysics (physics.geo-ph) #Machine Learning (cs.LG) #Reservoir Engineering and Simulation Methods

paper · pdf · doi:10.48550/arxiv.2109.03697

openalex publication_date 2021/09/03 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Numerical simulation of multiphase flow in porous media is essential for many geoscience applications. Machine learning models trained with numerical simulation data can provide a faster alternative to traditional simulators. Here we present U-FNO, a novel neural network architecture for solving multiphase flow problems with superior accuracy, speed, and data efficiency. U-FNO is designed based on the newly proposed Fourier neural operator (FNO), which has shown excellent performance in single-phase flows. We extend the FNO-based architecture to a highly complex CO2-water multiphase problem with wide ranges of permeability and porosity heterogeneity, anisotropy, reservoir conditions, injection configurations, flow rates, and multiphase flow properties. The U-FNO architecture is more accurate in gas saturation and pressure buildup predictions than the original FNO and a state-of-the-art convolutional neural network (CNN) benchmark. Meanwhile, it has superior data utilization efficiency, requiring only a third of the training data to achieve the equivalent accuracy as CNN. U-FNO provides superior performance in highly heterogeneous geological formations and critically important applications such as gas saturation and pressure buildup "fronts" determination. The trained model can serve as a general-purpose alternative to routine numerical simulations of 2D-radial CO2 injection problems with significant speed-ups than traditional simulators.

Cited by

• PGOT: A Physics-Geometry Operator Transformer for Complex PDEs
• PI-MFM: Physics-informed multimodal foundation model for solving partial differential equations
• Extending Fourier Neural Operators for Modeling Parameterized and Coupled PDEs
• Elliptic Regularity Theory in Barron Spaces and Applications to the Deep Ritz Method
• FLAME: Physics-Guided Neural Operators for Onboard Satellite Methane Detection in Hyperspectral Imagery
• Geometric Laplace Neural Operator
• Integrating Fourier Neural Operator with Diffusion Model for Autoregressive Predictions of Three-dimensional Turbulence
• Robust Variational Model Based Tailored UNet: Leveraging Edge Detector and Mean Curvature for Improved Image Segmentation
• MSPT: Efficient Large-Scale Physical Modeling via Parallelized Multi-Scale Attention
• Banach neural operator for Navier-Stokes equations
• Breaking Scale Anchoring: Frequency Representation Learning for Accurate High-Resolution Inference from Low-Resolution Training
• SAOT: An Enhanced Locality-Aware Spectral Transformer for Solving PDEs
• Learning from Imperfect Labels: A Physics-Aware Neural Operator with Application to DAS Data Denoising
• Fill the gaps: continuous in time interpolation of fluid dynamical simulations
• Emulating Radiative Transfer in Astrophysical Environments
• AgenticSciML: Collaborative Multi-Agent Systems for Emergent Discovery in Scientific Machine Learning
• Transolver is a Linear Transformer: Revisiting Physics-Attention through the Lens of Linear Attention
• SA-EMO: Structure-Aligned Encoder Mixture of Operators for Generalizable Full-waveform Inversion
• SSTODE: Ocean-Atmosphere Physics-Informed Neural ODEs for Sea Surface Temperature Prediction
• Hybrid DeepONet Surrogates for Multiphase Flow in Porous Media
• Finite Element Representation Network (FERN) for Operator Learning with a Localized Trainable Basis
• Incorporating Local Hölder Regularity into PINNs for Solving Elliptic PDEs
• Fourier Neural Operators for Two-Phase, 2D Mold-Filling Problems Related to Metal Casting
• EddyFormer: Accelerated Neural Simulations of Three-Dimensional Turbulence at Scale
• Accelerating Eigenvalue Dataset Generation via Chebyshev Subspace Filter
• MNO: Multiscale Neural Operator for Computational Fluid Dynamics with 3D Point Cloud Data
• Application of Reduced-Order Models for Temporal Multiscale Representations in the Prediction of Dynamical Systems
• Efficient High-Accuracy PDEs Solver with the Linear Attention Neural Operator
• Fourier Transform Multiple Instance Learning for Whole Slide Image Classification
• Self-Attention to Operator Learning-based 3D-IC Thermal Simulation
• A comprehensive comparison of neural operators for 3D industry-scale engineering designs
• Physics-informed Attention-enhanced Fourier Neural Operator for Solar Magnetic Field Extrapolations
• RheOFormer: A generative transformer model for simulation of complex fluids and flows
• PDE Solvers Should Be Local: Fast, Stable Rollouts with Learned Local Stencils
• DRIFT-Net: A Spectral--Coupled Neural Operator for PDEs Learning
• Multimodal Neural Operators for Real-Time Biomechanical Modelling of Traumatic Brain Injury
• Learning Greens Operators through Hierarchical Neural Networks Inspired by the Fast Multipole Method
• A reduced-order derivative-informed neural operator for subsurface fluid-flow
• ReBaNO: Reduced Basis Neural Operator Mitigating Generalization Gaps and Achieving Discretization Invariance
• Neural Network Solution of Non-Markovian Quantum State Diffusion and Operator Construction of Quantum Stochastic Process
• MD-PNOP: Equation-Recast Neural Operators for Minimal-Data Extrapolation and PDE Solver Acceleration
• VAE-DNN: Energy-Efficient Trainable-by-Parts Surrogate Model For Parametric Partial Differential Equations
• Physics-Constrained Fine-Tuning of Flow-Matching Models for Generation and Inverse Problems
• Spatiotemporal wall pressure forecast of a rectangular cylinder with physics-aware DeepU-Fourier neural network
• Light-Weight Diffusion Multiplier and Uncertainty Quantification for Fourier Neural Operators
• Modeling turbulent and self-gravitating fluids with Fourier neural operators

Related

• U-DeepONet: U-Net Enhanced Deep Operator Network for Geologic Carbon Sequestration
• Fourier-MIONet: Fourier-enhanced multiple-input neural operators for multiphase modeling of geological carbon sequestration
• Hybrid DeepONet Surrogates for Multiphase Flow in Porous Media
• A MgNO Method for Multiphase Flow in Porous Media
• A Physics-Constrained Deep Learning Model for Simulating Multiphase Flow in 3D Heterogeneous Porous Media