☆ 4.6 Article

A machine-learning supported multi-scale LBM-TPM model of unsaturated, anisotropic, and deformable porous materials

INTERNATIONAL JOURNAL FOR NUMERICAL AND ANALYTICAL METHODS IN GEOMECHANICS (2023)

Related references

Note: Only part of the references are listed.

Article Computer Science, Interdisciplinary Applications

A physics-informed convolutional neural network for the simulation and prediction of two-phase Darcy flows in heterogeneous porous media

Zhao Zhang et al.

Summary: The physics-informed convolutional neural network (PICNN) is developed for simulating transient two-phase Darcy flows in heterogeneous reservoir models with source/sink terms in the absence of labeled data. The finite volume method (FVM) is used to approximate the PDE residual in the loss function, and a well model is adopted to handle high pressure gradient near sources or sinks. The proposed approach is validated in homogeneous and heterogeneous reservoirs, and its efficiency and accuracy are discussed.

JOURNAL OF COMPUTATIONAL PHYSICS (2023)

Add to Collection

Article Engineering, Multidisciplinary

Deep CNNs as universal predictors of elasticity tensors in homogenization

Bernhard Eidel

Summary: In this work, 3D convolutional neural networks (CNNs) are trained to predict the elastic macroscale stiffness of random heterogeneous, multiphase materials. The proposed CNN model is universal and can handle various microstructures, phase fractions, and ranges of material properties. It predicts stiffness under different boundary conditions and provides an indicator for the representative volume element size. The model achieves high accuracy and greatly speeds up the prediction process compared to traditional methods.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2023)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Nanoscale soil-water retention mechanism of unsaturated clay via MD and machine learning

Zhe Zhang et al.

Summary: In this article, the nanoscale soil-water retention mechanism of unsaturated clay was investigated using molecular dynamics and machine learning. Pyrophyllite was selected as the precursor of other 2:1 clay minerals due to its stable structure. Molecular dynamics simulations of clay at low degrees of saturation were conducted, representing soil water as a point cloud using the center-of-mass method. The soil-water retention mechanism at the nanoscale was analyzed by distinguishing adsorptive pressure and capillary pressure at different mass water contents and considering the apparent capillary interface area. The results showed that the adsorption effect is dominated by van der Waals force and hydroxyl hydration, and capillarity plays a prominent role as the mass water content increases.

COMPUTERS AND GEOTECHNICS (2023)

Add to Collection

Article Materials Science, Multidisciplinary

Neural networks meet hyperelasticity: A guide to enforcing physics

Lennart Linden et al.

Summary: In this study, a hyperelastic constitutive model based on neural networks is proposed to apply to compressible material behavior. The model fulfills all common constitutive conditions, including symmetry, objectivity, material symmetry, polyconvexity, and thermodynamic consistency. The physically sensible stress behavior is ensured by using growth terms and normalization terms. By combining a sound mechanical basis with the flexibility of neural networks, the model harmonizes the theory of hyperelasticity with machine learning techniques. The non-negativity of the neural network-based potentials is examined numerically, and the model's applicability is demonstrated through data calibration and finite element simulations.

JOURNAL OF THE MECHANICS AND PHYSICS OF SOLIDS (2023)

Add to Collection

Article Materials Science, Multidisciplinary

Machine learning aided multiscale magnetostatics

Fadi Aldakheel et al.

Summary: Computational material modeling using CNN provides a solution for efficient and accurate modeling in heterogeneous materials, reducing the development costs and speeding up the design process.

MECHANICS OF MATERIALS (2023)

Add to Collection

Article Multidisciplinary Sciences

Quaternion to Euler angles conversion: A direct, general and computationally efficient method

Evandro Bernardes et al.

Summary: This paper presents a general formula for converting a rotation quaternion to Euler angles, which works with all possible sequences of rotations. It also provides a comparison with a popular matrix-to-Euler angle algorithm and demonstrates a faster execution speed of 30 times. The algorithm's conciseness and straightforward implementation make it highly appealing.

PLOS ONE (2022)

Add to Collection

Article Engineering, Geological

A multiscale LBM-TPM-PFM approach for modeling of multiphase fluid flow in fractured porous media

Mohamad Chaaban et al.

Summary: In this paper, a reliable micro-to-macroscale framework is presented to model multiphase fluid flow through fractured porous media. The lattice Boltzmann method (LBM) is utilized within the phase-field modeling (PFM) of fractures to achieve this. New phase-field-dependent relationships for various parameters are proposed and a multiscale concept for coupling is achieved. Numerical simulations on real microgeometries of fractured porous media validate the reliability of the model.

INTERNATIONAL JOURNAL FOR NUMERICAL AND ANALYTICAL METHODS IN GEOMECHANICS (2022)

Add to Collection

Article Energy & Fuels

An innovative application of deep learning in multiscale modeling of subsurface fluid flow: Reconstructing the basis functions of the mixed GMsFEM

Abouzar Choubineh et al.

Summary: This study investigates multiscale modeling by using data-driven methods to replace standard polynomial basis functions. Four distinct CNN models were developed to predict multiscale basis functions in mixed GMsFEM. The statistical results indicate that the AMSGrad optimization algorithm performs slightly better than Adam.

JOURNAL OF PETROLEUM SCIENCE AND ENGINEERING (2022)

Add to Collection

Article Engineering, Geological

An offline multi-scale unsaturated poromechanics model enabled by self-designed/self-improved neural networks

Yousef Heider et al.

Summary: This paper presents a meta-modeling approach that utilizes deep reinforcement learning to automatically discover optimal neural network settings for the machine learning constitutive laws. By replacing the human modeler to handle the optimized choices of setup, the AI agent self-learns from taking a sequence of actions within the selection environment. The resulting ML-generated material models can be integrated into a finite element solver to solve initial-boundary-value problems.

INTERNATIONAL JOURNAL FOR NUMERICAL AND ANALYTICAL METHODS IN GEOMECHANICS (2021)

Add to Collection

Article Engineering, Mechanical

1D convolutional neural networks and applications: A survey

Serkan Kiranyaz et al.

Summary: CNNs have become the standard for various Computer Vision and Machine Learning operations, with 1D CNNs offering advantages in applications with 1D signals. This paper provides a comprehensive review of 1D CNNs architecture, principles, and their recent progress in engineering applications.

MECHANICAL SYSTEMS AND SIGNAL PROCESSING (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

DNN2: A hyper-parameter reinforcement learning game for self-design of neural network based elasto-plastic constitutive descriptions

Alexander Fuchs et al.

Summary: This study introduces a meta-modeling framework using artificial intelligence to replicate the path-dependent constitutive responses of composite materials; a Deep Reinforcement Learning combinatorics game is invented to search for optimal hyper-parameter sets automatically; it explores the trade-off between different hyper-parameter configurations and the possibility of transferring hyper-parameter knowledge among different RVEs.

COMPUTERS & STRUCTURES (2021)

Add to Collection

Review Mechanics

A review on phase-field modeling of hydraulic fracturing

Yousef Heider

Summary: This paper reviews the significant scientific works utilizing PFM to model hydraulic fracturing since 2012, covering different phase-field evolution formulations, stiffness degradation functions, crack driving forces, and methods for estimating crack aperture width. It also discusses various numerical approaches and implementations, concluding with open topics and challenges for future research.

ENGINEERING FRACTURE MECHANICS (2021)

Add to Collection

Article Geosciences, Multidisciplinary

Estimating Dispersion Coefficient in Flow Through Heterogeneous Porous Media by a Deep Convolutional Neural Network

Serveh Kamrava et al.

Summary: The combination of deep convolutional neural network (DCNN) and random-walk particle-tracking (RWPT) simulation provides a powerful tool for rapidly and accurately estimating the longitudinal dispersion coefficient DL in flow through porous media, allowing for the study of flow-related phenomena and properties in geological formations.

GEOPHYSICAL RESEARCH LETTERS (2021)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Deep multiscale model learning

Yating Wang et al.

JOURNAL OF COMPUTATIONAL PHYSICS (2020)

Add to Collection

Article Engineering, Multidisciplinary

SO(3)-invariance of informed-graph-based deep neural network for anisotropic elastoplastic materials

Yousef Heider et al.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2020)

Add to Collection

Article Thermodynamics

Upscaling LBM-TPM simulation approach of Darcy and non-Darcy fluid flow in deformable, heterogeneous porous media

Mohamad Chaaban et al.

INTERNATIONAL JOURNAL OF HEAT AND FLUID FLOW (2020)

Add to Collection

Article Multidisciplinary Sciences

Machine Learning Coupled Multi-Scale Modeling for Redox Flow Batteries

Jie Bao et al.

ADVANCED THEORY AND SIMULATIONS (2020)

Add to Collection

Article Engineering, Multidisciplinary

An updated Lagrangian LBM-DEM-FEM coupling model for dual-permeability fissured porous media with embedded discontinuities

Kun Wang et al.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2019)

Add to Collection

Article Engineering, Multidisciplinary

A graph theoretic framework for representation, exploration and analysis on computed states of physical systems

R. Banerjee et al.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2019)

Add to Collection

Article Engineering, Multidisciplinary

A multiscale multi-permeability poroplasticity model linked by recursive homogenizations and deep learning

Kun Wang et al.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2018)

Add to Collection

Article Acoustics

Deep convolutional neural networks for estimating porous material parameters with ultrasound tomography

Timo Lahivaara et al.

JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA (2018)

Add to Collection

Article Computer Science, Artificial Intelligence

Recent advances in convolutional neural networks

Jiuxiang Gu et al.

PATTERN RECOGNITION (2018)

Add to Collection

Article Environmental Sciences

An estimation of the main wetting branch of the soil water retention curve based on its main drying branch using the machine learning method

Krzysztof Lamorski et al.

WATER RESOURCES RESEARCH (2017)

Add to Collection

Article Engineering, Multidisciplinary

A semi-implicit discrete-continuum coupling method for porous media based on the effective stress principle at finite strain

Kun Wang et al.

COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING (2016)

Add to Collection

Article Engineering, Geological

Stress-induced anisotropy in granular materials: fabric, stiffness, and permeability

Matthew R. Kuhn et al.

ACTA GEOTECHNICA (2015)

Add to Collection

Article Computer Science, Interdisciplinary Applications

Digital rock physics benchmarks-Part I: Imaging and segmentation

Heiko Andrae et al.

COMPUTERS & GEOSCIENCES (2013)

Add to Collection

Article Physics, Fluids & Plasmas

Numerical study on the permeability in a tensorial form for laminar flow in anisotropic porous media

S. A. Galindo-Torres et al.

PHYSICAL REVIEW E (2012)

Add to Collection

Article Geosciences, Multidisciplinary

Connecting microstructural attributes and permeability from 3D tomographic images of in situ shear-enhanced compaction bands using multiscale computations

WaiChing Sun et al.

GEOPHYSICAL RESEARCH LETTERS (2011)

Add to Collection

Article Engineering, Multidisciplinary

Comparison of monolithic and splitting solution schemes for dynamic porous media problems

B. Markert et al.

INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN ENGINEERING (2010)

Add to Collection

Article Engineering, Chemical

A constitutive approach to 3-d nonlinear fluid flow through finite deformable porous continua - With application to a high-porosity polyurethane foam

Bernd Markert

TRANSPORT IN POROUS MEDIA (2007)

Add to Collection

Article Mathematics, Applied

Explicit jump immersed interface method for virtual material design of the effective elastic moduli of composite materials

Vita Rutka et al.

NUMERICAL ALGORITHMS (2006)

Add to Collection

Article Environmental Sciences

Lattice-Boltzmann simulation of two-phase flow in porous media

C Pan et al.

WATER RESOURCES RESEARCH (2004)

Add to Collection

Article Physics, Mathematical

Lattice-Boltzmann simulation of capillary rise dynamics

P Raiskinmäki et al.

JOURNAL OF STATISTICAL PHYSICS (2002)

Add to Collection

Article Materials Science, Multidisciplinary

Spreading dynamics of three-dimensional droplets by the lattice-Boltzmann method

P Raiskinmäki et al.

COMPUTATIONAL MATERIALS SCIENCE (2000)

Add to Collection

© Peeref 2019-2024. All rights reserved.