No Cover Image

Journal article 189 views 79 downloads

Deep Learning Predicts Stress–Strain Relations of Granular Materials Based on Triaxial Testing Data

Tongming QU, Shaocheng Di, Yuntian Feng Orcid Logo, Min Wang, Tingting Zhao, Mengqi Wang

Computer Modeling in Engineering & Sciences, Volume: 128, Issue: 1, Pages: 129 - 144

Swansea University Authors: Tongming QU, Yuntian Feng Orcid Logo

  • 57281.pdf

    PDF | Version of Record

    This work is licensed under a Creative Commons Attribution 4.0 International License

    Download (1.24MB)

Abstract

This study presents an AI-based constitutive modelling framework wherein the prediction model directly learns from triaxial testing data by combining discrete element modelling (DEM) and deep learning. A constitutive learning strategy is proposed based on the generally accepted frame-indifference as...

Full description

Published in: Computer Modeling in Engineering & Sciences
ISSN: 1526-1506
Published: Computers, Materials and Continua (Tech Science Press) 2021
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa57281
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2021-07-08T08:08:18Z
last_indexed 2021-07-31T03:16:14Z
id cronfa57281
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2021-07-30T11:46:39.2817459</datestamp><bib-version>v2</bib-version><id>57281</id><entry>2021-07-08</entry><title>Deep Learning Predicts Stress&#x2013;Strain Relations of Granular Materials Based on Triaxial Testing Data</title><swanseaauthors><author><sid>1a8144ef1058bc1310206808a4d274c3</sid><firstname>Tongming</firstname><surname>QU</surname><name>Tongming QU</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>d66794f9c1357969a5badf654f960275</sid><ORCID>0000-0002-6396-8698</ORCID><firstname>Yuntian</firstname><surname>Feng</surname><name>Yuntian Feng</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2021-07-08</date><deptcode>FGSEN</deptcode><abstract>This study presents an AI-based constitutive modelling framework wherein the prediction model directly learns from triaxial testing data by combining discrete element modelling (DEM) and deep learning. A constitutive learning strategy is proposed based on the generally accepted frame-indifference assumption in constructing material constitutive models. The low-dimensional principal stress-strain sequence pairs, measured from discrete element modelling of triaxial testing, are used to train recurrent neural networks, and then the predicted principal stress sequence is augmented to other high-dimensional or general stress tensor via coordinate transformation. Through detailed hyperparameter investigations, it is found that long short-term memory (LSTM) and gated recurrent unit (GRU) networks have similar prediction performance in constitutive modelling problems, and both satisfactorily predict the stress responses of granular materials subjected to a given unseen strain path. Furthermore, the unique merits and ongoing challenges of data-driven constitutive models for granular materials are discussed.</abstract><type>Journal Article</type><journal>Computer Modeling in Engineering &amp; Sciences</journal><volume>128</volume><journalNumber>1</journalNumber><paginationStart>129</paginationStart><paginationEnd>144</paginationEnd><publisher>Computers, Materials and Continua (Tech Science Press)</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>1526-1506</issnPrint><issnElectronic/><keywords>Deep learning; granular materials; constitutive modelling; discrete element modelling; coordinate transformation; LSTM; GRU</keywords><publishedDay>28</publishedDay><publishedMonth>6</publishedMonth><publishedYear>2021</publishedYear><publishedDate>2021-06-28</publishedDate><doi>10.32604/cmes.2021.016172</doi><url/><notes/><college>COLLEGE NANME</college><department>Science and Engineering - Faculty</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>FGSEN</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2021-07-30T11:46:39.2817459</lastEdited><Created>2021-07-08T09:00:57.5983050</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Civil Engineering</level></path><authors><author><firstname>Tongming</firstname><surname>QU</surname><order>1</order></author><author><firstname>Shaocheng</firstname><surname>Di</surname><order>2</order></author><author><firstname>Yuntian</firstname><surname>Feng</surname><orcid>0000-0002-6396-8698</orcid><order>3</order></author><author><firstname>Min</firstname><surname>Wang</surname><order>4</order></author><author><firstname>Tingting</firstname><surname>Zhao</surname><order>5</order></author><author><firstname>Mengqi</firstname><surname>Wang</surname><order>6</order></author></authors><documents><document><filename>57281__20358__433ff06393a14f03b8aff58d484a56d2.pdf</filename><originalFilename>57281.pdf</originalFilename><uploaded>2021-07-08T09:07:54.8604137</uploaded><type>Output</type><contentLength>1295588</contentLength><contentType>application/pdf</contentType><version>Version of Record</version><cronfaStatus>true</cronfaStatus><documentNotes>This work is licensed under a Creative Commons Attribution 4.0 International License</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>http://creativecommons.org/licenses/by/4.0/</licence></document></documents><OutputDurs/></rfc1807>
spelling 2021-07-30T11:46:39.2817459 v2 57281 2021-07-08 Deep Learning Predicts Stress–Strain Relations of Granular Materials Based on Triaxial Testing Data 1a8144ef1058bc1310206808a4d274c3 Tongming QU Tongming QU true false d66794f9c1357969a5badf654f960275 0000-0002-6396-8698 Yuntian Feng Yuntian Feng true false 2021-07-08 FGSEN This study presents an AI-based constitutive modelling framework wherein the prediction model directly learns from triaxial testing data by combining discrete element modelling (DEM) and deep learning. A constitutive learning strategy is proposed based on the generally accepted frame-indifference assumption in constructing material constitutive models. The low-dimensional principal stress-strain sequence pairs, measured from discrete element modelling of triaxial testing, are used to train recurrent neural networks, and then the predicted principal stress sequence is augmented to other high-dimensional or general stress tensor via coordinate transformation. Through detailed hyperparameter investigations, it is found that long short-term memory (LSTM) and gated recurrent unit (GRU) networks have similar prediction performance in constitutive modelling problems, and both satisfactorily predict the stress responses of granular materials subjected to a given unseen strain path. Furthermore, the unique merits and ongoing challenges of data-driven constitutive models for granular materials are discussed. Journal Article Computer Modeling in Engineering & Sciences 128 1 129 144 Computers, Materials and Continua (Tech Science Press) 1526-1506 Deep learning; granular materials; constitutive modelling; discrete element modelling; coordinate transformation; LSTM; GRU 28 6 2021 2021-06-28 10.32604/cmes.2021.016172 COLLEGE NANME Science and Engineering - Faculty COLLEGE CODE FGSEN Swansea University 2021-07-30T11:46:39.2817459 2021-07-08T09:00:57.5983050 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Civil Engineering Tongming QU 1 Shaocheng Di 2 Yuntian Feng 0000-0002-6396-8698 3 Min Wang 4 Tingting Zhao 5 Mengqi Wang 6 57281__20358__433ff06393a14f03b8aff58d484a56d2.pdf 57281.pdf 2021-07-08T09:07:54.8604137 Output 1295588 application/pdf Version of Record true This work is licensed under a Creative Commons Attribution 4.0 International License true eng http://creativecommons.org/licenses/by/4.0/
title Deep Learning Predicts Stress–Strain Relations of Granular Materials Based on Triaxial Testing Data
spellingShingle Deep Learning Predicts Stress–Strain Relations of Granular Materials Based on Triaxial Testing Data
Tongming QU
Yuntian Feng
title_short Deep Learning Predicts Stress–Strain Relations of Granular Materials Based on Triaxial Testing Data
title_full Deep Learning Predicts Stress–Strain Relations of Granular Materials Based on Triaxial Testing Data
title_fullStr Deep Learning Predicts Stress–Strain Relations of Granular Materials Based on Triaxial Testing Data
title_full_unstemmed Deep Learning Predicts Stress–Strain Relations of Granular Materials Based on Triaxial Testing Data
title_sort Deep Learning Predicts Stress–Strain Relations of Granular Materials Based on Triaxial Testing Data
author_id_str_mv 1a8144ef1058bc1310206808a4d274c3
d66794f9c1357969a5badf654f960275
author_id_fullname_str_mv 1a8144ef1058bc1310206808a4d274c3_***_Tongming QU
d66794f9c1357969a5badf654f960275_***_Yuntian Feng
author Tongming QU
Yuntian Feng
author2 Tongming QU
Shaocheng Di
Yuntian Feng
Min Wang
Tingting Zhao
Mengqi Wang
format Journal article
container_title Computer Modeling in Engineering & Sciences
container_volume 128
container_issue 1
container_start_page 129
publishDate 2021
institution Swansea University
issn 1526-1506
doi_str_mv 10.32604/cmes.2021.016172
publisher Computers, Materials and Continua (Tech Science Press)
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Civil Engineering{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Civil Engineering
document_store_str 1
active_str 0
description This study presents an AI-based constitutive modelling framework wherein the prediction model directly learns from triaxial testing data by combining discrete element modelling (DEM) and deep learning. A constitutive learning strategy is proposed based on the generally accepted frame-indifference assumption in constructing material constitutive models. The low-dimensional principal stress-strain sequence pairs, measured from discrete element modelling of triaxial testing, are used to train recurrent neural networks, and then the predicted principal stress sequence is augmented to other high-dimensional or general stress tensor via coordinate transformation. Through detailed hyperparameter investigations, it is found that long short-term memory (LSTM) and gated recurrent unit (GRU) networks have similar prediction performance in constitutive modelling problems, and both satisfactorily predict the stress responses of granular materials subjected to a given unseen strain path. Furthermore, the unique merits and ongoing challenges of data-driven constitutive models for granular materials are discussed.
published_date 2021-06-28T04:05:52Z
_version_ 1756237662525587456
score 10.926594