No Cover Image

Journal article 63 views

Digital twinning of thermal systems: A comparison between supervised and reinforcement learning

Armando Di Meglio, Nicola Massarotti, Perumal Nithiarasu Orcid Logo

Computational Thermal Sciences: An International Journal, Volume: 17, Issue: 3, Pages: 39 - 46

Swansea University Author: Perumal Nithiarasu Orcid Logo

  • Accepted Manuscript under embargo until: 18th February 2026

Check full text

DOI (Published version): 10.1615/computthermalscien.2025057435

Abstract

This article explores two novel approaches for controlling heat transfer systems through the development of “digital twins”, focusing on transient thermal systems. The study involves creating a digital representation of a physical system, specifically a 2D square subjected to an inward and transient...

Full description

Published in: Computational Thermal Sciences: An International Journal
ISSN: 1940-2503
Published: Begell House 2025
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa68929
first_indexed 2025-02-20T09:00:28Z
last_indexed 2025-03-26T05:31:12Z
id cronfa68929
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2025-03-25T10:53:42.4575435</datestamp><bib-version>v2</bib-version><id>68929</id><entry>2025-02-20</entry><title>Digital twinning of thermal systems: A comparison between supervised and reinforcement learning</title><swanseaauthors><author><sid>3b28bf59358fc2b9bd9a46897dbfc92d</sid><ORCID>0000-0002-4901-2980</ORCID><firstname>Perumal</firstname><surname>Nithiarasu</surname><name>Perumal Nithiarasu</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2025-02-20</date><deptcode>ACEM</deptcode><abstract>This article explores two novel approaches for controlling heat transfer systems through the development of &#x201C;digital twins&#x201D;, focusing on transient thermal systems. The study involves creating a digital representation of a physical system, specifically a 2D square subjected to an inward and transient heat flux, with the goal of keeping the maximum temperature within a predefined limit by changing the convective cooling. The first method utilizes a neural network trained on steady-state data, whereas the second employs an interactive learning algorithm. Results show that both strategies prove to be effective in managing the system's thermal performance. However, the RL-based approach demonstrates greater flexibility in adapting to new scenarios, albeit at the cost of increased computational demands due to the necessity of integrating interactive learning with unsteady Finite Element Method (FEM) simulations for training, validation, and testing phases.</abstract><type>Journal Article</type><journal>Computational Thermal Sciences: An International Journal</journal><volume>17</volume><journalNumber>3</journalNumber><paginationStart>39</paginationStart><paginationEnd>46</paginationEnd><publisher>Begell House</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>1940-2503</issnPrint><issnElectronic/><keywords>digital twin, heat transfer, interactive AI, deep learning, machine learning, FEM simulation</keywords><publishedDay>18</publishedDay><publishedMonth>2</publishedMonth><publishedYear>2025</publishedYear><publishedDate>2025-02-18</publishedDate><doi>10.1615/computthermalscien.2025057435</doi><url/><notes/><college>COLLEGE NANME</college><department>Aerospace, Civil, Electrical, and Mechanical Engineering</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>ACEM</DepartmentCode><institution>Swansea University</institution><apcterm/><funders/><projectreference/><lastEdited>2025-03-25T10:53:42.4575435</lastEdited><Created>2025-02-20T08:55:18.0485987</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Electronic and Electrical Engineering</level></path><authors><author><firstname>Armando Di</firstname><surname>Meglio</surname><order>1</order></author><author><firstname>Nicola</firstname><surname>Massarotti</surname><order>2</order></author><author><firstname>Perumal</firstname><surname>Nithiarasu</surname><orcid>0000-0002-4901-2980</orcid><order>3</order></author></authors><documents><document><filename>Under embargo</filename><originalFilename>Under embargo</originalFilename><uploaded>2025-02-20T09:00:17.7131945</uploaded><type>Output</type><contentLength>532779</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><embargoDate>2026-02-18T00:00:00.0000000</embargoDate><copyrightCorrect>true</copyrightCorrect><language>eng</language></document></documents><OutputDurs/></rfc1807>
spelling 2025-03-25T10:53:42.4575435 v2 68929 2025-02-20 Digital twinning of thermal systems: A comparison between supervised and reinforcement learning 3b28bf59358fc2b9bd9a46897dbfc92d 0000-0002-4901-2980 Perumal Nithiarasu Perumal Nithiarasu true false 2025-02-20 ACEM This article explores two novel approaches for controlling heat transfer systems through the development of “digital twins”, focusing on transient thermal systems. The study involves creating a digital representation of a physical system, specifically a 2D square subjected to an inward and transient heat flux, with the goal of keeping the maximum temperature within a predefined limit by changing the convective cooling. The first method utilizes a neural network trained on steady-state data, whereas the second employs an interactive learning algorithm. Results show that both strategies prove to be effective in managing the system's thermal performance. However, the RL-based approach demonstrates greater flexibility in adapting to new scenarios, albeit at the cost of increased computational demands due to the necessity of integrating interactive learning with unsteady Finite Element Method (FEM) simulations for training, validation, and testing phases. Journal Article Computational Thermal Sciences: An International Journal 17 3 39 46 Begell House 1940-2503 digital twin, heat transfer, interactive AI, deep learning, machine learning, FEM simulation 18 2 2025 2025-02-18 10.1615/computthermalscien.2025057435 COLLEGE NANME Aerospace, Civil, Electrical, and Mechanical Engineering COLLEGE CODE ACEM Swansea University 2025-03-25T10:53:42.4575435 2025-02-20T08:55:18.0485987 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Electronic and Electrical Engineering Armando Di Meglio 1 Nicola Massarotti 2 Perumal Nithiarasu 0000-0002-4901-2980 3 Under embargo Under embargo 2025-02-20T09:00:17.7131945 Output 532779 application/pdf Accepted Manuscript true 2026-02-18T00:00:00.0000000 true eng
title Digital twinning of thermal systems: A comparison between supervised and reinforcement learning
spellingShingle Digital twinning of thermal systems: A comparison between supervised and reinforcement learning
Perumal Nithiarasu
title_short Digital twinning of thermal systems: A comparison between supervised and reinforcement learning
title_full Digital twinning of thermal systems: A comparison between supervised and reinforcement learning
title_fullStr Digital twinning of thermal systems: A comparison between supervised and reinforcement learning
title_full_unstemmed Digital twinning of thermal systems: A comparison between supervised and reinforcement learning
title_sort Digital twinning of thermal systems: A comparison between supervised and reinforcement learning
author_id_str_mv 3b28bf59358fc2b9bd9a46897dbfc92d
author_id_fullname_str_mv 3b28bf59358fc2b9bd9a46897dbfc92d_***_Perumal Nithiarasu
author Perumal Nithiarasu
author2 Armando Di Meglio
Nicola Massarotti
Perumal Nithiarasu
format Journal article
container_title Computational Thermal Sciences: An International Journal
container_volume 17
container_issue 3
container_start_page 39
publishDate 2025
institution Swansea University
issn 1940-2503
doi_str_mv 10.1615/computthermalscien.2025057435
publisher Begell House
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 - Electronic and Electrical Engineering{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Electronic and Electrical Engineering
document_store_str 0
active_str 0
description This article explores two novel approaches for controlling heat transfer systems through the development of “digital twins”, focusing on transient thermal systems. The study involves creating a digital representation of a physical system, specifically a 2D square subjected to an inward and transient heat flux, with the goal of keeping the maximum temperature within a predefined limit by changing the convective cooling. The first method utilizes a neural network trained on steady-state data, whereas the second employs an interactive learning algorithm. Results show that both strategies prove to be effective in managing the system's thermal performance. However, the RL-based approach demonstrates greater flexibility in adapting to new scenarios, albeit at the cost of increased computational demands due to the necessity of integrating interactive learning with unsteady Finite Element Method (FEM) simulations for training, validation, and testing phases.
published_date 2025-02-18T09:31:57Z
_version_ 1829003766972547072
score 11.058203

Similar Items