No Cover Image

Journal article 487 views 327 downloads

Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning

Xiang Li, Zhanli Liu, Shaoqing Cui, Chengcheng Luo, Chenfeng Li Orcid Logo, Zhuo Zhuang

Computer Methods in Applied Mechanics and Engineering, Volume: 347, Pages: 735 - 753

Swansea University Author: Chenfeng Li Orcid Logo

Abstract

In contrast to the composition uniformity of homogeneous materials, heterogeneous materials are normally composed of two or more distinctive constituents. It is usually recognized that the effective material property of a heterogeneous material is related to the mechanical property and the distribut...

Full description

Published in: Computer Methods in Applied Mechanics and Engineering
ISSN: 00457825
Published: 2019
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa48290
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2019-01-18T20:01:16Z
last_indexed 2019-03-11T14:00:16Z
id cronfa48290
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2019-03-11T11:09:06.8953298</datestamp><bib-version>v2</bib-version><id>48290</id><entry>2019-01-18</entry><title>Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning</title><swanseaauthors><author><sid>82fe170d5ae2c840e538a36209e5a3ac</sid><ORCID>0000-0003-0441-211X</ORCID><firstname>Chenfeng</firstname><surname>Li</surname><name>Chenfeng Li</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2019-01-18</date><deptcode>CIVL</deptcode><abstract>In contrast to the composition uniformity of homogeneous materials, heterogeneous materials are normally composed of two or more distinctive constituents. It is usually recognized that the effective material property of a heterogeneous material is related to the mechanical property and the distribution pattern of each forming constituent. However, to establish an explicit relationship between the macroscale mechanical property and the microstructure appears to be complicated. On the other hand, machine learning methods are broadly employed to excavate inherent rules and correlations based on a significant amount of data samples. Specifically, deep neural networks are established to deal with situations where input&#x2013;output mappings are extensively complex. In this paper, a method is proposed to establish the implicit mapping between the effective mechanical property and the mesoscale structure of heterogeneous materials. Shale is employed in this paper as an example to illustrate the method. At the mesoscale, a shale sample is a complex heterogeneous composite that consists of multiple mineral constituents. The mechanical properties of each mineral constituent vary significantly, and mineral constituents are distributed in an utterly random manner within shale samples. Large quantities of shale samples are generated based on mesoscale scanning electron microscopy images using a stochastic reconstruction algorithm. Image processing techniques are employed to transform the shale sample images to finite element models. Finite element analysis is utilized to evaluate the effective mechanical properties of the shale samples. A convolutional neural network is trained based on the images of stochastic shale samples and their effective moduli. The trained network is validated to be able to predict the effective moduli of real shale samples accurately and efficiently. Not limited to shale, the proposed method can be further extended to predict effective mechanical properties of other heterogeneous materials.</abstract><type>Journal Article</type><journal>Computer Methods in Applied Mechanics and Engineering</journal><volume>347</volume><paginationStart>735</paginationStart><paginationEnd>753</paginationEnd><publisher/><issnPrint>00457825</issnPrint><keywords>Heterogeneous materials, Shale, Deep learning, Stochastic reconstruction</keywords><publishedDay>31</publishedDay><publishedMonth>12</publishedMonth><publishedYear>2019</publishedYear><publishedDate>2019-12-31</publishedDate><doi>10.1016/j.cma.2019.01.005</doi><url/><notes/><college>COLLEGE NANME</college><department>Civil Engineering</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>CIVL</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2019-03-11T11:09:06.8953298</lastEdited><Created>2019-01-18T16:24:22.0305128</Created><path><level id="1">College of Engineering</level><level id="2">Engineering</level></path><authors><author><firstname>Xiang</firstname><surname>Li</surname><order>1</order></author><author><firstname>Zhanli</firstname><surname>Liu</surname><order>2</order></author><author><firstname>Shaoqing</firstname><surname>Cui</surname><order>3</order></author><author><firstname>Chengcheng</firstname><surname>Luo</surname><order>4</order></author><author><firstname>Chenfeng</firstname><surname>Li</surname><orcid>0000-0003-0441-211X</orcid><order>5</order></author><author><firstname>Zhuo</firstname><surname>Zhuang</surname><order>6</order></author></authors><documents><document><filename>0048290-18012019162711.pdf</filename><originalFilename>li2019(2)v2.pdf</originalFilename><uploaded>2019-01-18T16:27:11.7000000</uploaded><type>Output</type><contentLength>11389287</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><embargoDate>2020-01-14T00:00:00.0000000</embargoDate><copyrightCorrect>true</copyrightCorrect><language>eng</language></document></documents><OutputDurs/></rfc1807>
spelling 2019-03-11T11:09:06.8953298 v2 48290 2019-01-18 Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning 82fe170d5ae2c840e538a36209e5a3ac 0000-0003-0441-211X Chenfeng Li Chenfeng Li true false 2019-01-18 CIVL In contrast to the composition uniformity of homogeneous materials, heterogeneous materials are normally composed of two or more distinctive constituents. It is usually recognized that the effective material property of a heterogeneous material is related to the mechanical property and the distribution pattern of each forming constituent. However, to establish an explicit relationship between the macroscale mechanical property and the microstructure appears to be complicated. On the other hand, machine learning methods are broadly employed to excavate inherent rules and correlations based on a significant amount of data samples. Specifically, deep neural networks are established to deal with situations where input–output mappings are extensively complex. In this paper, a method is proposed to establish the implicit mapping between the effective mechanical property and the mesoscale structure of heterogeneous materials. Shale is employed in this paper as an example to illustrate the method. At the mesoscale, a shale sample is a complex heterogeneous composite that consists of multiple mineral constituents. The mechanical properties of each mineral constituent vary significantly, and mineral constituents are distributed in an utterly random manner within shale samples. Large quantities of shale samples are generated based on mesoscale scanning electron microscopy images using a stochastic reconstruction algorithm. Image processing techniques are employed to transform the shale sample images to finite element models. Finite element analysis is utilized to evaluate the effective mechanical properties of the shale samples. A convolutional neural network is trained based on the images of stochastic shale samples and their effective moduli. The trained network is validated to be able to predict the effective moduli of real shale samples accurately and efficiently. Not limited to shale, the proposed method can be further extended to predict effective mechanical properties of other heterogeneous materials. Journal Article Computer Methods in Applied Mechanics and Engineering 347 735 753 00457825 Heterogeneous materials, Shale, Deep learning, Stochastic reconstruction 31 12 2019 2019-12-31 10.1016/j.cma.2019.01.005 COLLEGE NANME Civil Engineering COLLEGE CODE CIVL Swansea University 2019-03-11T11:09:06.8953298 2019-01-18T16:24:22.0305128 College of Engineering Engineering Xiang Li 1 Zhanli Liu 2 Shaoqing Cui 3 Chengcheng Luo 4 Chenfeng Li 0000-0003-0441-211X 5 Zhuo Zhuang 6 0048290-18012019162711.pdf li2019(2)v2.pdf 2019-01-18T16:27:11.7000000 Output 11389287 application/pdf Accepted Manuscript true 2020-01-14T00:00:00.0000000 true eng
title Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning
spellingShingle Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning
Chenfeng Li
title_short Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning
title_full Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning
title_fullStr Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning
title_full_unstemmed Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning
title_sort Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning
author_id_str_mv 82fe170d5ae2c840e538a36209e5a3ac
author_id_fullname_str_mv 82fe170d5ae2c840e538a36209e5a3ac_***_Chenfeng Li
author Chenfeng Li
author2 Xiang Li
Zhanli Liu
Shaoqing Cui
Chengcheng Luo
Chenfeng Li
Zhuo Zhuang
format Journal article
container_title Computer Methods in Applied Mechanics and Engineering
container_volume 347
container_start_page 735
publishDate 2019
institution Swansea University
issn 00457825
doi_str_mv 10.1016/j.cma.2019.01.005
college_str College of Engineering
hierarchytype
hierarchy_top_id collegeofengineering
hierarchy_top_title College of Engineering
hierarchy_parent_id collegeofengineering
hierarchy_parent_title College of Engineering
department_str Engineering{{{_:::_}}}College of Engineering{{{_:::_}}}Engineering
document_store_str 1
active_str 0
description In contrast to the composition uniformity of homogeneous materials, heterogeneous materials are normally composed of two or more distinctive constituents. It is usually recognized that the effective material property of a heterogeneous material is related to the mechanical property and the distribution pattern of each forming constituent. However, to establish an explicit relationship between the macroscale mechanical property and the microstructure appears to be complicated. On the other hand, machine learning methods are broadly employed to excavate inherent rules and correlations based on a significant amount of data samples. Specifically, deep neural networks are established to deal with situations where input–output mappings are extensively complex. In this paper, a method is proposed to establish the implicit mapping between the effective mechanical property and the mesoscale structure of heterogeneous materials. Shale is employed in this paper as an example to illustrate the method. At the mesoscale, a shale sample is a complex heterogeneous composite that consists of multiple mineral constituents. The mechanical properties of each mineral constituent vary significantly, and mineral constituents are distributed in an utterly random manner within shale samples. Large quantities of shale samples are generated based on mesoscale scanning electron microscopy images using a stochastic reconstruction algorithm. Image processing techniques are employed to transform the shale sample images to finite element models. Finite element analysis is utilized to evaluate the effective mechanical properties of the shale samples. A convolutional neural network is trained based on the images of stochastic shale samples and their effective moduli. The trained network is validated to be able to predict the effective moduli of real shale samples accurately and efficiently. Not limited to shale, the proposed method can be further extended to predict effective mechanical properties of other heterogeneous materials.
published_date 2019-12-31T04:01:01Z
_version_ 1737027025824120832
score 10.898023