☆ 4.7 Article

Rapid Permeability Upscaling of Digital Porous Media via Physics-Informed Neural Networks

WATER RESOURCES RESEARCH (2023)

Related references

Note: Only part of the references are listed.

Article Multidisciplinary Sciences

Physics-informed dynamic mode decomposition

Peter J. J. Baddoo et al.

Summary: In this work, the integration of physical principles into the dynamic mode decomposition (DMD) is demonstrated. A physics-informed DMD (piDMD) optimization is proposed to restrict the models to a matrix manifold that respects the physical structure of the system. Several closed-form solutions and efficient algorithms for the corresponding piDMD optimizations are derived based on fundamental physical principles. The piDMD models outperform standard DMD algorithms in various applications, showing advantages in spectral identification, state prediction, and estimation.

PROCEEDINGS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES (2023)

Add to Collection

Article Engineering, Multidisciplinary

Improved training of physics-informed neural networks for parabolic differential equations with sharply perturbed initial conditions

Yifei Zong et al.

Summary: We propose a multi-component approach to improve the training of the physics-informed neural network (PINN) model for parabolic problems with sharply perturbed initial conditions. By considering the advection-dispersion equation (ADE) with a point (Gaussian) source initial condition, we find that the decay rate of perturbations in the initial condition can lead to large approximation errors in the PINN solution for d & GE; 2. In addition, localized large gradients in the ADE solution make the Latin hypercube sampling of the equation's residual, which is common in PINN, highly inefficient. We also demonstrate the sensitivity of the PINN solution of parabolic equations to the choice of weights in the loss function.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2023)

Add to Collection

Article Energy & Fuels

A CNN-based approach for upscaling multiphase flow in digital sandstones

Javad Siavashi et al.

Summary: The study utilizes convolutional neural networks and downsampling techniques to train high-resolution images for predicting and assigning upscaled properties to low-resolution models, resulting in satisfactory outcomes in multiphase flow simulations.

FUEL (2022)

Add to Collection

Article Energy & Fuels

Influence of Heterogeneity on Carbonate Permeability Upscaling: A Renormalization Approach Coupled with the Pore Network Model

Haiyang Zhang et al.

Summary: This article modifies an established method for permeability upscaling in carbonate rocks and proposes a new regression-based approach. The results show that the new method provides more accurate permeability estimations for full-size core plug computations and is superior to the classical method for small-scale sample estimations. These findings contribute to the understanding of permeability upscaling and the impact of heterogeneity in carbonate rocks.

ENERGY & FUELS (2022)

Add to Collection

Article Environmental Sciences

Generalizable Permeability Prediction of Digital Porous Media via a Novel Multi-Scale 3D Convolutional Neural Network

Mohamed Elmorsy et al.

Summary: Subsurface characterization is crucial for understanding and controlling underground processes, but current numerical simulation methods are computationally expensive. Machine learning models offer a more efficient approach and have potential for predicting subsurface permeability accurately.

WATER RESOURCES RESEARCH (2022)

Add to Collection

Article Mathematics, Applied

Scientific Machine Learning Through Physics-Informed Neural Networks: Where we are and What's Next

Salvatore Cuomo et al.

Summary: Physics-Informed Neural Networks (PINN) are a type of neural network that incorporates model equations, such as partial differential equations, as a component. PINNs have been used to solve various types of equations, including fractional equations and stochastic partial differential equations. Current research focuses on optimizing PINN through different aspects, such as activation functions, gradient optimization techniques, neural network structures, and loss function structures. Despite the demonstrated feasibility of PINN in certain cases compared to traditional numerical techniques, there are still unresolved theoretical issues.

JOURNAL OF SCIENTIFIC COMPUTING (2022)

Add to Collection

Article Energy & Fuels

Multiscale Digital Rock Imaging and Modeling for Measuring the Heterogeneous Carbonate and Conglomerate Permeability at the Laboratory Plug Scale

Qianghui Xu et al.

Summary: This study developed multiscale workflows to predict the permeability of fractured carbonates, vuggy carbonates, and conglomerates. By comparing different methods, the best permeability analysis method for these three rock types was identified.

ENERGY & FUELS (2022)

Add to Collection

Review Engineering, Mechanical

Physics-informed neural networks (PINNs) for fluid mechanics: a review

Shengze Cai et al.

Summary: Significant progress has been made in simulating flow problems over the last 50 years, but challenges remain in incorporating noisy data, complex mesh generation, and solving high-dimensional problems. Physics-informed neural networks (PINNs) have been demonstrated as effective in solving inverse flow problems related to various fluid dynamics scenarios.

ACTA MECHANICA SINICA (2021)

Add to Collection

Article Mechanics

A point-cloud deep learning framework for prediction of fluid flow fields on irregular geometries

Ali Kashefi et al.

Summary: The study introduces a novel deep learning framework for flow field predictions in irregular domains, using a neural network based on the PointNet architecture to map between spatial positions and CFD quantities. The network inherits desirable features of unstructured meshes, accurately represents object geometry, and preserves the accuracy of CFD data without data interpolation. The new approach outperforms existing methods in terms of speed and accuracy.

PHYSICS OF FLUIDS (2021)

Add to Collection

Article Multidisciplinary Sciences

Upscaling the porosity-permeability relationship of a microporous carbonate for Darcy-scale flow with machine learning

H. P. Menke et al.

Summary: The permeability of pore structures is crucial in predicting large-scale flow behavior, especially for complex rocks with multimodal porosity structures. Recent advances in machine learning combined with numerical modelling and structural analysis provide deeper insights into the relationship between structure and permeability.

SCIENTIFIC REPORTS (2021)

Add to Collection

Article Multidisciplinary Sciences

Machine learning-accelerated computational fluid dynamics

Dmitrii Kochkov et al.

Summary: Numerical simulation of fluids is crucial in modeling physical phenomena like weather and aerodynamics, but solving the Navier-Stokes equations at scale remains challenging due to computational costs. This study uses end-to-end deep learning to improve approximations in computational fluid dynamics for modeling two-dimensional turbulent flows, achieving computational speedups and stability while generalizing well to different scenarios. Leveraging machine learning and hardware accelerators, this approach showcases the potential to enhance simulations without sacrificing accuracy or generalization.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2021)

Add to Collection

Article Mathematics, Applied

Learning dynamical systems from data: A simple cross-validation perspective, part I: Parametric kernel flows

Boumediene Hamzi et al.

Summary: Regressing the vector field of a dynamical system from a finite number of observed states is a natural way to learn surrogate models for such systems. Variants of cross-validation, such as Kernel Flows, can be used as simple approaches for learning the kernel used in these emulators.

PHYSICA D-NONLINEAR PHENOMENA (2021)

Add to Collection

Article Environmental Sciences

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

QiZhi He et al.

Summary: A discretization-free approach based on the PINN method is proposed to solve the coupled advection-dispersion equation and Darcy flow equation, using deep neural networks for approximating hydraulic conductivity, hydraulic head, and concentration fields, leading to improved accuracy in computations. The PINN method demonstrates high accuracy for both forward and backward ADEs, with the inclusion of concentration measurements significantly enhancing solution accuracy.

WATER RESOURCES RESEARCH (2021)

Add to Collection

Article Chemistry, Physical

Simulating fluid flow in complex porous materials by integrating the governing equations with deep-layered machines

Serveh Kamrava et al.

Summary: A machine learning approach incorporating mass conservation and the Navier-Stokes equations is introduced for accurately predicting flow properties of porous media with reduced computation time. The method shows high accuracy in predictions and can be applied to different porous media with reliable results.

NPJ COMPUTATIONAL MATERIALS (2021)

Add to Collection

Review Physics, Applied

Physics-informed machine learning

George Em Karniadakis et al.

Summary: Physics-informed learning seamlessly integrates data and mathematical models through neural networks or kernel-based regression networks for accurate inference of realistic and high-dimensional multiphysics problems. Challenges remain in incorporating noisy data seamlessly, complex mesh generation, and addressing high-dimensional problems.

NATURE REVIEWS PHYSICS (2021)

Add to Collection

Review Mechanics

Machine Learning for Fluid Mechanics

Steven L. Brunton et al.

ANNUAL REVIEW OF FLUID MECHANICS, VOL 52 (2020)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Modeling the dynamics of PDE systems with physics-constrained deep auto-regressive networks

Nicholas Geneva et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2020)

Add to Collection

Article Engineering, Multidisciplinary

Machine learning in cardiovascular flows modeling: Predicting arterial blood pressure from non-invasive 4D flow MRI data using physics-informed neural networks

Georgios Kissas et al.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2020)

Add to Collection

Article Water Resources

Machine learning in geo- and environmental sciences: From small to large scale

Pejman Tahmasebi et al.

ADVANCES IN WATER RESOURCES (2020)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Kernel Flows: From learning kernels from data into the abyss

Houman Owhadi et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2019)

Add to Collection

Article Water Resources

Computations of permeability of large rock images by dual grid domain decomposition

Ying Da Wang et al.

ADVANCES IN WATER RESOURCES (2019)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Physics-constrained deep learning for high-dimensional surrogate modeling and uncertainty quantification without labeled data

Yinhao Zhu et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2019)

Add to Collection

Article Energy & Fuels

Application of the renormalization group approach for permeability estimation in digital rocks

Shuaishuai Wei et al.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2019)

Add to Collection

Article Computer Science, Theory & Methods

A survey on Image Data Augmentation for Deep Learning

Connor Shorten et al.

JOURNAL OF BIG DATA (2019)

Add to Collection

Article Computer Science, Theory & Methods

Survey on deep learning with class imbalance

Justin M. Johnson et al.

JOURNAL OF BIG DATA (2019)

Add to Collection

Article Computer Science, Artificial Intelligence

SMOTE for Learning from Imbalanced Data: Progress and Challenges, Marking the 15-year Anniversary

Alberto Fernandez et al.

JOURNAL OF ARTIFICIAL INTELLIGENCE RESEARCH (2018)

Add to Collection

Article Energy & Fuels

Scale up of pore-scale transport properties from micro to macro scale; network modelling approach

Farzad Bashtani et al.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2018)

Add to Collection

Article Medical Informatics

The effects of varying class distribution on learner behavior for medicare fraud detection with imbalanced big data

Richard A. Bauder et al.

HEALTH INFORMATION SCIENCE AND SYSTEMS (2018)

Add to Collection

Article Energy & Fuels

Industrial applications of digital rock technology

Carl Fredrik Berg et al.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2017)

Add to Collection

Article Water Resources

The impact of porous media heterogeneity on non-Darcy flow behaviour from pore-scale simulation

Bagus P. Muljadi et al.

ADVANCES IN WATER RESOURCES (2016)

Add to Collection

Article Mathematics, Interdisciplinary Applications

HYBRID MULTISCALE FINITE VOLUME METHOD FOR ADVECTION-DIFFUSION EQUATIONS SUBJECT TO HETEROGENEOUS REACTIVE BOUNDARY CONDITIONS

David A. Barajas-Solano et al.

MULTISCALE MODELING & SIMULATION (2016)

Add to Collection

Article Computer Science, Interdisciplinary Applications

An open-source toolbox for multiphase flow in porous media

P. Horgue et al.

COMPUTER PHYSICS COMMUNICATIONS (2015)

Add to Collection

Review Multidisciplinary Sciences

Deep learning

Yann LeCun et al.

NATURE (2015)

Add to Collection

Article Computer Science, Artificial Intelligence

Resampling strategies for regression

Luis Torgo et al.

EXPERT SYSTEMS (2015)

Add to Collection

Article Water Resources

Pore-scale imaging and modelling

Martin J. Blunt et al.

ADVANCES IN WATER RESOURCES (2013)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Digital rock physics benchmarks-part II: Computing effective properties

Heiko Andrae et al.

COMPUTERS & GEOSCIENCES (2013)

Add to Collection

Article Geosciences, Multidisciplinary

Computations of Absolute Permeability on Micro-CT Images

Peyman Mostaghimi et al.

MATHEMATICAL GEOSCIENCES (2013)

Add to Collection

Article Multidisciplinary Sciences

Lifetime of carbon capture and storage as a climate-change mitigation technology

Michael L. Szulczewski et al.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2012)

Add to Collection

Article Mathematics, Applied

Lattice-Boltzmann studies of fluid flow in porous media with realistic rock geometries

Edo S. Boek et al.

COMPUTERS & MATHEMATICS WITH APPLICATIONS (2010)

Add to Collection

Article Engineering, Chemical

New Renormalization Schemes for Conductivity Upscaling in Heterogeneous Media

M. R. Karim et al.

TRANSPORT IN POROUS MEDIA (2010)

Add to Collection

Article Physics, Fluids & Plasmas

Pore-network extraction from micro-computerized-tomography images

Hu Dong et al.

PHYSICAL REVIEW E (2009)

Add to Collection

Article Engineering, Chemical

Analytical three-dimensional renormalization for calculating effective permeabilities

Christopher P. Green et al.

TRANSPORT IN POROUS MEDIA (2007)

Add to Collection

© Peeref 2019-2024. All rights reserved.