No Cover Image

E-Thesis 68 views 64 downloads

AI Mesh Informed Techniques for Optimising the Design Process / CALLUM LOCK

Swansea University Author: CALLUM LOCK

DOI (Published version): 10.23889/SUThesis.71072

Abstract

This thesis presents a novel, data-driven framework for automatically generating near-optimal unstructured meshes for computational simulations. The primary objective is to reduce the manual effort and expert intervention typically required in mesh generation by leveraging historical simulation data...

Full description

Published: Swansea 2025
Institution: Swansea University
Degree level: Doctoral
Degree name: Ph.D
Supervisor: Hassan, O.; Sevilla, R.; and Jones, J.
URI: https://cronfa.swan.ac.uk/Record/cronfa71072
first_indexed 2025-12-04T09:58:53Z
last_indexed 2025-12-05T18:13:22Z
id cronfa71072
recordtype RisThesis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2025-12-04T10:23:49.2677804</datestamp><bib-version>v2</bib-version><id>71072</id><entry>2025-12-04</entry><title>AI Mesh Informed Techniques for Optimising the Design Process</title><swanseaauthors><author><sid>f311bd38eefcadf983a7d99c8058e942</sid><firstname>CALLUM</firstname><surname>LOCK</surname><name>CALLUM LOCK</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2025-12-04</date><abstract>This thesis presents a novel, data-driven framework for automatically generating near-optimal unstructured meshes for computational simulations. The primary objective is to reduce the manual e&#xFB00;ort and expert intervention typically required in mesh generation by leveraging historical simulation data and arti&#xFB01;cial neural networks (ANNs) to predict appropriate mesh spacing &#xFB01;elds. The work is motivated by the growing availability of high-&#xFB01;delity simulation data in industry and the need to streamline simulation work&#xFB02;ows &#x2013; particularly in the aerospace sector, where the mesh generation process remains one of the most resource-intensive steps.Three di&#xFB00;erent strategies are developed and evaluated. The &#xFB01;rst approach predicts the properties of point sources used to de&#xFB01;ne the mesh resolution. The second introduces a coarse background mesh, onto which spacing functions are conservatively interpolated and predicted by ANNs. The third and &#xFB01;nal approach extends the method to fully anisotropic spacing by predicting the components of the metric tensor, allowing for directionally aligned mesh re&#xFB01;nement. All three techniques are trained on datasets de-rived from prior simulations and are shown to generalise e&#xFB00;ectively to unseen geometric and &#xFB02;ow conditions.Extensive numerical experiments in three-dimensional compressible &#xFB02;ow scenarios &#x2013;including wings and full aircraft con&#xFB01;gurations, demonstrate that the proposed methods yield high-quality meshes capable of producing accurate solutions. Furthermore, an environmental impact analysis shows the potential for a substantial reduction in computational cost and energy usage, highlighting the ability of the methods outlined to be part of sustainable simulation practices.This work lays the foundation for integrating machine learning into the meshing pipeline, enabling intelligent, scalable, and more e&#xFB03;cient simulation-driven design across a wide range of engineering applications.</abstract><type>E-Thesis</type><journal/><volume/><journalNumber/><paginationStart/><paginationEnd/><publisher/><placeOfPublication>Swansea</placeOfPublication><isbnPrint/><isbnElectronic/><issnPrint/><issnElectronic/><keywords>Mesh generation, Machine learning, Near-optimal mesh prediction, Computational fluiddynamics</keywords><publishedDay>28</publishedDay><publishedMonth>8</publishedMonth><publishedYear>2025</publishedYear><publishedDate>2025-08-28</publishedDate><doi>10.23889/SUThesis.71072</doi><url/><notes/><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><supervisor>Hassan, O.; Sevilla, R.; and Jones, J.</supervisor><degreelevel>Doctoral</degreelevel><degreename>Ph.D</degreename><degreesponsorsfunders>EPSRC, Airbus Defence and Space</degreesponsorsfunders><apcterm/><funders>EPSRC, Airbus Defence and Space</funders><projectreference/><lastEdited>2025-12-04T10:23:49.2677804</lastEdited><Created>2025-12-04T09:55:05.1383065</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Aerospace Engineering</level></path><authors><author><firstname>CALLUM</firstname><surname>LOCK</surname><order>1</order></author></authors><documents><document><filename>71072__35751__4a15cca2600b411c9014c19d1ad45dd0.pdf</filename><originalFilename>2025_Lock_C.final.71072.pdf</originalFilename><uploaded>2025-12-04T10:02:36.9165156</uploaded><type>Output</type><contentLength>65677952</contentLength><contentType>application/pdf</contentType><version>E-Thesis &#x2013; open access</version><cronfaStatus>true</cronfaStatus><documentNotes>Copyright: the author, Callum Lock, 2025</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language></document></documents><OutputDurs/></rfc1807>
spelling 2025-12-04T10:23:49.2677804 v2 71072 2025-12-04 AI Mesh Informed Techniques for Optimising the Design Process f311bd38eefcadf983a7d99c8058e942 CALLUM LOCK CALLUM LOCK true false 2025-12-04 This thesis presents a novel, data-driven framework for automatically generating near-optimal unstructured meshes for computational simulations. The primary objective is to reduce the manual effort and expert intervention typically required in mesh generation by leveraging historical simulation data and artificial neural networks (ANNs) to predict appropriate mesh spacing fields. The work is motivated by the growing availability of high-fidelity simulation data in industry and the need to streamline simulation workflows – particularly in the aerospace sector, where the mesh generation process remains one of the most resource-intensive steps.Three different strategies are developed and evaluated. The first approach predicts the properties of point sources used to define the mesh resolution. The second introduces a coarse background mesh, onto which spacing functions are conservatively interpolated and predicted by ANNs. The third and final approach extends the method to fully anisotropic spacing by predicting the components of the metric tensor, allowing for directionally aligned mesh refinement. All three techniques are trained on datasets de-rived from prior simulations and are shown to generalise effectively to unseen geometric and flow conditions.Extensive numerical experiments in three-dimensional compressible flow scenarios –including wings and full aircraft configurations, demonstrate that the proposed methods yield high-quality meshes capable of producing accurate solutions. Furthermore, an environmental impact analysis shows the potential for a substantial reduction in computational cost and energy usage, highlighting the ability of the methods outlined to be part of sustainable simulation practices.This work lays the foundation for integrating machine learning into the meshing pipeline, enabling intelligent, scalable, and more efficient simulation-driven design across a wide range of engineering applications. E-Thesis Swansea Mesh generation, Machine learning, Near-optimal mesh prediction, Computational fluiddynamics 28 8 2025 2025-08-28 10.23889/SUThesis.71072 COLLEGE NANME COLLEGE CODE Swansea University Hassan, O.; Sevilla, R.; and Jones, J. Doctoral Ph.D EPSRC, Airbus Defence and Space EPSRC, Airbus Defence and Space 2025-12-04T10:23:49.2677804 2025-12-04T09:55:05.1383065 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Aerospace Engineering CALLUM LOCK 1 71072__35751__4a15cca2600b411c9014c19d1ad45dd0.pdf 2025_Lock_C.final.71072.pdf 2025-12-04T10:02:36.9165156 Output 65677952 application/pdf E-Thesis – open access true Copyright: the author, Callum Lock, 2025 true eng
title AI Mesh Informed Techniques for Optimising the Design Process
spellingShingle AI Mesh Informed Techniques for Optimising the Design Process
CALLUM LOCK
title_short AI Mesh Informed Techniques for Optimising the Design Process
title_full AI Mesh Informed Techniques for Optimising the Design Process
title_fullStr AI Mesh Informed Techniques for Optimising the Design Process
title_full_unstemmed AI Mesh Informed Techniques for Optimising the Design Process
title_sort AI Mesh Informed Techniques for Optimising the Design Process
author_id_str_mv f311bd38eefcadf983a7d99c8058e942
author_id_fullname_str_mv f311bd38eefcadf983a7d99c8058e942_***_CALLUM LOCK
author CALLUM LOCK
author2 CALLUM LOCK
format E-Thesis
publishDate 2025
institution Swansea University
doi_str_mv 10.23889/SUThesis.71072
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 - Aerospace Engineering{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Aerospace Engineering
document_store_str 1
active_str 0
description This thesis presents a novel, data-driven framework for automatically generating near-optimal unstructured meshes for computational simulations. The primary objective is to reduce the manual effort and expert intervention typically required in mesh generation by leveraging historical simulation data and artificial neural networks (ANNs) to predict appropriate mesh spacing fields. The work is motivated by the growing availability of high-fidelity simulation data in industry and the need to streamline simulation workflows – particularly in the aerospace sector, where the mesh generation process remains one of the most resource-intensive steps.Three different strategies are developed and evaluated. The first approach predicts the properties of point sources used to define the mesh resolution. The second introduces a coarse background mesh, onto which spacing functions are conservatively interpolated and predicted by ANNs. The third and final approach extends the method to fully anisotropic spacing by predicting the components of the metric tensor, allowing for directionally aligned mesh refinement. All three techniques are trained on datasets de-rived from prior simulations and are shown to generalise effectively to unseen geometric and flow conditions.Extensive numerical experiments in three-dimensional compressible flow scenarios –including wings and full aircraft configurations, demonstrate that the proposed methods yield high-quality meshes capable of producing accurate solutions. Furthermore, an environmental impact analysis shows the potential for a substantial reduction in computational cost and energy usage, highlighting the ability of the methods outlined to be part of sustainable simulation practices.This work lays the foundation for integrating machine learning into the meshing pipeline, enabling intelligent, scalable, and more efficient simulation-driven design across a wide range of engineering applications.
published_date 2025-08-28T05:27:59Z
_version_ 1851641435119419392
score 11.089905

Similar Items