No Cover Image

Journal article 365 views

Data-driven multiscale modelling of granular materials via knowledge transfer and sharing

Tongming Qu Orcid Logo, Jidong Zhao, Shaoheng Guan Orcid Logo, Yuntian Feng Orcid Logo

International Journal of Plasticity, Volume: 171, Start page: 103786

Swansea University Author: Yuntian Feng Orcid Logo

Full text not available from this repository: check for access using links below.

Check full text

DOI (Published version): 10.1016/j.ijplas.2023.103786

Abstract

Machine learning approaches have found immense potential to revolutionise the constitutive modelling of granular materials. However, data scarcity poses a significant challenge to this emerging paradigm. This study aims to tackle this issue by presenting two transfer learning-based strategies that h...

Full description

Published in: International Journal of Plasticity
ISSN: 0749-6419
Published: Elsevier BV 2023
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa64899
Tags: Add Tag
No Tags, Be the first to tag this record!
first_indexed 2023-11-02T08:40:01Z
last_indexed 2023-11-02T08:40:01Z
id cronfa64899
recordtype SURis
fullrecord <?xml version="1.0" encoding="utf-8"?><rfc1807 xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xmlns:xsd="http://www.w3.org/2001/XMLSchema"><bib-version>v2</bib-version><id>64899</id><entry>2023-11-02</entry><title>Data-driven multiscale modelling of granular materials via knowledge transfer and sharing</title><swanseaauthors><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>2023-11-02</date><deptcode>CIVL</deptcode><abstract>Machine learning approaches have found immense potential to revolutionise the constitutive modelling of granular materials. However, data scarcity poses a significant challenge to this emerging paradigm. This study aims to tackle this issue by presenting two transfer learning-based strategies that harness well-established constitutive knowledge and similar material data to reduce data demands for data-driven material modelling. The first approach utilises phenomenological constitutive models to generate massive synthetic data which reflect the targeted material behaviour to train a base model. This base model is then repurposed for a new task based on numerical simulation data via transfer learning. The other approach involves using available material data to train a base model, which is then applied to other new materials that are similar but with limited data. The proposed transfer learning methods are tested on both particle-scale simulations of representative volume elements (RVEs) and hierarchical multiscale modelling of boundary value problems (BVPs) of granular materials. The trained data-driven material model is embedded in numerical simulations with the finite element method (FEM) to validate its accuracy, efficiency, and stability. The results demonstrate that transfer learning can effectively achieve high-quality machine learning predictions with limited data. The transfer learning strategy presented in this study is expected to be widely applicable to small data-driven material modelling.</abstract><type>Journal Article</type><journal>International Journal of Plasticity</journal><volume>171</volume><journalNumber/><paginationStart>103786</paginationStart><paginationEnd/><publisher>Elsevier BV</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0749-6419</issnPrint><issnElectronic/><keywords>Granular materials; DEM; Machine learning; Transfer learning; Data-driven material modelling; Hierarchical multiscale modelling</keywords><publishedDay>1</publishedDay><publishedMonth>12</publishedMonth><publishedYear>2023</publishedYear><publishedDate>2023-12-01</publishedDate><doi>10.1016/j.ijplas.2023.103786</doi><url/><notes/><college>COLLEGE NANME</college><department>Civil Engineering</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>CIVL</DepartmentCode><institution>Swansea University</institution><apcterm>Other</apcterm><funders>The study was financially supported by the National Natural Science Foundation of China (via General Project #11972030) and the Research Grants Council of Hong Kong (under GRF #16208720).</funders><projectreference/><lastEdited>2024-03-08T15:44:15.3012359</lastEdited><Created>2023-11-02T08:38:10.0343543</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><orcid>0000-0003-3058-8282</orcid><order>1</order></author><author><firstname>Jidong</firstname><surname>Zhao</surname><order>2</order></author><author><firstname>Shaoheng</firstname><surname>Guan</surname><orcid>0000-0001-7867-9517</orcid><order>3</order></author><author><firstname>Yuntian</firstname><surname>Feng</surname><orcid>0000-0002-6396-8698</orcid><order>4</order></author></authors><documents/><OutputDurs/></rfc1807>
spelling v2 64899 2023-11-02 Data-driven multiscale modelling of granular materials via knowledge transfer and sharing d66794f9c1357969a5badf654f960275 0000-0002-6396-8698 Yuntian Feng Yuntian Feng true false 2023-11-02 CIVL Machine learning approaches have found immense potential to revolutionise the constitutive modelling of granular materials. However, data scarcity poses a significant challenge to this emerging paradigm. This study aims to tackle this issue by presenting two transfer learning-based strategies that harness well-established constitutive knowledge and similar material data to reduce data demands for data-driven material modelling. The first approach utilises phenomenological constitutive models to generate massive synthetic data which reflect the targeted material behaviour to train a base model. This base model is then repurposed for a new task based on numerical simulation data via transfer learning. The other approach involves using available material data to train a base model, which is then applied to other new materials that are similar but with limited data. The proposed transfer learning methods are tested on both particle-scale simulations of representative volume elements (RVEs) and hierarchical multiscale modelling of boundary value problems (BVPs) of granular materials. The trained data-driven material model is embedded in numerical simulations with the finite element method (FEM) to validate its accuracy, efficiency, and stability. The results demonstrate that transfer learning can effectively achieve high-quality machine learning predictions with limited data. The transfer learning strategy presented in this study is expected to be widely applicable to small data-driven material modelling. Journal Article International Journal of Plasticity 171 103786 Elsevier BV 0749-6419 Granular materials; DEM; Machine learning; Transfer learning; Data-driven material modelling; Hierarchical multiscale modelling 1 12 2023 2023-12-01 10.1016/j.ijplas.2023.103786 COLLEGE NANME Civil Engineering COLLEGE CODE CIVL Swansea University Other The study was financially supported by the National Natural Science Foundation of China (via General Project #11972030) and the Research Grants Council of Hong Kong (under GRF #16208720). 2024-03-08T15:44:15.3012359 2023-11-02T08:38:10.0343543 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Civil Engineering Tongming Qu 0000-0003-3058-8282 1 Jidong Zhao 2 Shaoheng Guan 0000-0001-7867-9517 3 Yuntian Feng 0000-0002-6396-8698 4
title Data-driven multiscale modelling of granular materials via knowledge transfer and sharing
spellingShingle Data-driven multiscale modelling of granular materials via knowledge transfer and sharing
Yuntian Feng
title_short Data-driven multiscale modelling of granular materials via knowledge transfer and sharing
title_full Data-driven multiscale modelling of granular materials via knowledge transfer and sharing
title_fullStr Data-driven multiscale modelling of granular materials via knowledge transfer and sharing
title_full_unstemmed Data-driven multiscale modelling of granular materials via knowledge transfer and sharing
title_sort Data-driven multiscale modelling of granular materials via knowledge transfer and sharing
author_id_str_mv d66794f9c1357969a5badf654f960275
author_id_fullname_str_mv d66794f9c1357969a5badf654f960275_***_Yuntian Feng
author Yuntian Feng
author2 Tongming Qu
Jidong Zhao
Shaoheng Guan
Yuntian Feng
format Journal article
container_title International Journal of Plasticity
container_volume 171
container_start_page 103786
publishDate 2023
institution Swansea University
issn 0749-6419
doi_str_mv 10.1016/j.ijplas.2023.103786
publisher Elsevier BV
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 0
active_str 0
description Machine learning approaches have found immense potential to revolutionise the constitutive modelling of granular materials. However, data scarcity poses a significant challenge to this emerging paradigm. This study aims to tackle this issue by presenting two transfer learning-based strategies that harness well-established constitutive knowledge and similar material data to reduce data demands for data-driven material modelling. The first approach utilises phenomenological constitutive models to generate massive synthetic data which reflect the targeted material behaviour to train a base model. This base model is then repurposed for a new task based on numerical simulation data via transfer learning. The other approach involves using available material data to train a base model, which is then applied to other new materials that are similar but with limited data. The proposed transfer learning methods are tested on both particle-scale simulations of representative volume elements (RVEs) and hierarchical multiscale modelling of boundary value problems (BVPs) of granular materials. The trained data-driven material model is embedded in numerical simulations with the finite element method (FEM) to validate its accuracy, efficiency, and stability. The results demonstrate that transfer learning can effectively achieve high-quality machine learning predictions with limited data. The transfer learning strategy presented in this study is expected to be widely applicable to small data-driven material modelling.
published_date 2023-12-01T15:44:11Z
_version_ 1792973368530567168
score 11.036706

Similar Items