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

Choubineh, Abouzar; Chen, Jie; Coenen, Frans; Ma, Fei

doi:10.1016/j.petrol.2022.110751

Cited by 3 publications

(2 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As a solution for the shortage of observed data, they employed computational data for training the ML model. In multiscale modeling of subsurface fluid flow in heterogeneous porous media, Choubineh et al 22 developed distinct convolutional neural networks (CNNs) to predict different multiscale basis functions for the mixed generalized multiscale finite element method. The solely required input for their models in the study was the permeability property.…”

Section: Introductionmentioning

confidence: 99%

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

Chaaban,

Heider,

Sun

et al. 2023

Num Anal Meth Geomechanics

View full text Add to dashboard Cite

The purpose of this paper is to investigate the utilization of artificial neural networks (ANNs) in learning models that address the nonlinear anisotropic flow and hysteresis retention behavior of deformable porous materials. Herein, the micro‐geometries of various networks of porous Bentheimer Sandstones subjected to several degrees of strain from the literature are considered. For the generation of the database required for the training, validation, and testing of the machine learning (ML) models, single‐phase and biphasic lattice Boltzmann (LB) simulations are performed. The anisotropic nature of the intrinsic permeability is investigated for the single‐phase LB simulations. Thereafter, the database contains the computed average fluid velocities versus the pressure gradients. In this database, the range of applied fluid pressure gradients includes Darcy as well as non‐Darcy flows. The generated output from the single‐phase flow simulations is implemented in a feed‐forward neural network, representing a path‐independent informed graph‐based model. Concerning the two‐phase LB simulations, the Shan‐Chen multiphase LB model is used to generate the retention curves of the cyclic drying/wetting processes in the deformed porous networks. Consequently, two different ML path‐dependent approaches, that is, 1D convolutional neural network and the recurrent neural network, are used to model the biphasic flow through the deformable porous materials. A comparison in terms of accuracy and speed of training between the two approaches is presented. Conclusively, the outcomes of the papers show the capability of the ML models in representing constitutive relations for permeability and hysteretic retention curves accurately and efficiently.

show abstract

Section: Introductionmentioning

confidence: 99%

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

Chaaban,

Heider,

Sun

et al. 2023

Num Anal Meth Geomechanics

View full text Add to dashboard Cite

show abstract

“…With maximum child development, children can focus on the learning process so that the objectives of learning can be achieved. One part of the learning media that can be used in the learning process in line with the times is the Canva application (Choubineh et al, 2022;Zhao, 2021). Canva is an online-based application that includes templates and various interesting features, such as logos, posters, presentations, graphics, brochures, videos and so on.…”

Section: Introductionmentioning

confidence: 99%

Canva Learning Media to Improve Students’ Learning Achievement in the Merdeka Curriculum

Prihantini,

Istianti,

Rukmini

et al. 2023

ijmst

View full text Add to dashboard Cite

Changes and developments in technology have brought the world of education involved in the utilization of these technologies. In today's digital era, it will never be separated from how efforts to improve the performance of the world of education by utilizing technology and applications. One of the most popular applications in the current disruption era that can be used to support educational operations is Canva. Canva is an online-based application that provides design templates and attractive features. The purpose of this study is to describe learning methods using Canva applications effectively enough to improve student learning achievement in the independent curriculum. This research uses quantitative methods, data obtained through distributing questionnaires. The questionnaire was distributed using the Google Form application. The results of this study explain that the use of Canva as a learning media can explore the creativity side of educators and improve student achievement in the teaching and learning process, create interesting learning materials so that they can arouse student interest in learning and lead to better learning outcomes. The conclusion of this study explains that Canva learning media greatly influences the process and learning outcomes of students in the independent curriculum. The limitations in this study are that researchers have difficulty in deepening the data, the time of data collection which takes a long time, and high subjectivity. This study also recommends that future researchers be able to conduct further and in-depth studies on the utilization of this canva learning media.

show abstract

Fourier Neural Operator for Fluid Flow in Small-Shape 2D Simulated Porous Media Dataset

et al. 2023

View full text Add to dashboard Cite

Machine Learning (ML) and/or Deep Learning (DL) methods can be used to predict fluid flow in porous media, as a suitable replacement for classical numerical approaches. Such data-driven approaches attempt to learn mappings between finite-dimensional Euclidean spaces. A novel neural framework, named Fourier Neural Operator (FNO), has been recently developed to act on infinite-dimensional spaces. A high proportion of the research available on the FNO has focused on problems with large-shape data. Furthermore, most published studies apply the FNO method to existing datasets. This paper applies and evaluates FNO to predict pressure distribution over a small, specified shape-data problem using 1700 Finite Element Method (FEM) generated samples, from heterogeneous permeability fields as the input. Considering FEM-calculated outputs as the true values, the configured FNO model provides superior prediction performance to that of a Convolutional Neural Network (CNN) in terms of statistical error assessment based on the coefficient of determination (R2) and Mean Squared Error (MSE). Sensitivity analysis considering a range of FNO configurations reveals that the most accurate model is obtained using modes=15 and width=100. Graphically, the FNO model precisely follows the observed trend in each porous medium evaluated. There is potential to further improve the FNO’s performance by including physics constraints in its network configuration.

show abstract

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

Cited by 3 publications

References 28 publications

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

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

Canva Learning Media to Improve Students’ Learning Achievement in the Merdeka Curriculum

Fourier Neural Operator for Fluid Flow in Small-Shape 2D Simulated Porous Media Dataset

Contact Info

Product

Resources

About