Reconstruction of Three-Dimensional Porous Media Using Deep Transfer Learning

The reconstruction of porous media is widely used in the study of fluid flows and engineering sciences. Some traditional reconstruction methods for porous media use the features extracted from real natural porous media and copy them to realize reconstructions. Currently, as one of the important bran...

Full description

Saved in:
Bibliographic Details
Main Authors: Yi Du, Jie Chen, Ting Zhang
Format: Article
Language:English
Published: Wiley 2020-01-01
Series:Geofluids
Online Access:http://dx.doi.org/10.1155/2020/6641642
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832552514621276160
author Yi Du
Jie Chen
Ting Zhang
author_facet Yi Du
Jie Chen
Ting Zhang
author_sort Yi Du
collection DOAJ
description The reconstruction of porous media is widely used in the study of fluid flows and engineering sciences. Some traditional reconstruction methods for porous media use the features extracted from real natural porous media and copy them to realize reconstructions. Currently, as one of the important branches of machine learning methods, the deep transfer learning (DTL) method has shown good performance in extracting features and transferring them to the predicted objects, which can be used for the reconstruction of porous media. Hence, a method for reconstructing porous media is presented by applying DTL to extract features from a training image (TI) of porous media to replace the process of scanning a TI for different patterns as in multiple-point statistical methods. The deep neural network is practically used to extract the complex features from the TI of porous media, and then, a reconstructed result can be obtained by transfer learning through copying these features. The proposed method was evaluated on shale and sandstone samples by comparing multiple-point connectivity functions, variogram curves, permeability, porosity, etc. The experimental results show that the proposed method is of high efficiency while preserving similar features with the target image, shortening reconstruction time, and reducing the burdens on CPU.
format Article
id doaj-art-8a300389abf344c5a51e325b50fcfc61
institution Kabale University
issn 1468-8115
1468-8123
language English
publishDate 2020-01-01
publisher Wiley
record_format Article
series Geofluids
spelling doaj-art-8a300389abf344c5a51e325b50fcfc612025-02-03T05:58:27ZengWileyGeofluids1468-81151468-81232020-01-01202010.1155/2020/66416426641642Reconstruction of Three-Dimensional Porous Media Using Deep Transfer LearningYi Du0Jie Chen1Ting Zhang2College of Engineering, Shanghai Polytechnic University, Shanghai 201209, ChinaCollege of Computer Science and Technology, Shanghai University of Electric Power, Shanghai 200090, ChinaCollege of Computer Science and Technology, Shanghai University of Electric Power, Shanghai 200090, ChinaThe reconstruction of porous media is widely used in the study of fluid flows and engineering sciences. Some traditional reconstruction methods for porous media use the features extracted from real natural porous media and copy them to realize reconstructions. Currently, as one of the important branches of machine learning methods, the deep transfer learning (DTL) method has shown good performance in extracting features and transferring them to the predicted objects, which can be used for the reconstruction of porous media. Hence, a method for reconstructing porous media is presented by applying DTL to extract features from a training image (TI) of porous media to replace the process of scanning a TI for different patterns as in multiple-point statistical methods. The deep neural network is practically used to extract the complex features from the TI of porous media, and then, a reconstructed result can be obtained by transfer learning through copying these features. The proposed method was evaluated on shale and sandstone samples by comparing multiple-point connectivity functions, variogram curves, permeability, porosity, etc. The experimental results show that the proposed method is of high efficiency while preserving similar features with the target image, shortening reconstruction time, and reducing the burdens on CPU.http://dx.doi.org/10.1155/2020/6641642
spellingShingle Yi Du
Jie Chen
Ting Zhang
Reconstruction of Three-Dimensional Porous Media Using Deep Transfer Learning
Geofluids
title Reconstruction of Three-Dimensional Porous Media Using Deep Transfer Learning
title_full Reconstruction of Three-Dimensional Porous Media Using Deep Transfer Learning
title_fullStr Reconstruction of Three-Dimensional Porous Media Using Deep Transfer Learning
title_full_unstemmed Reconstruction of Three-Dimensional Porous Media Using Deep Transfer Learning
title_short Reconstruction of Three-Dimensional Porous Media Using Deep Transfer Learning
title_sort reconstruction of three dimensional porous media using deep transfer learning
url http://dx.doi.org/10.1155/2020/6641642
work_keys_str_mv AT yidu reconstructionofthreedimensionalporousmediausingdeeptransferlearning
AT jiechen reconstructionofthreedimensionalporousmediausingdeeptransferlearning
AT tingzhang reconstructionofthreedimensionalporousmediausingdeeptransferlearning