No Cover Image

Journal article 423 views 130 downloads

Graph network simulators (GNS) for modelling particle-based fluid flow with a given inlet velocity

Philip Pe, Rajesh Ransing Orcid Logo

International Journal of Numerical Methods for Heat & Fluid Flow, Volume: 35, Issue: 9, Pages: 3053 - 3079

Swansea University Authors: Philip Pe, Rajesh Ransing Orcid Logo

  • particle_trickle_algorithm - AAM.pdf

    PDF | Accepted Manuscript

    This author accepted manuscript is deposited under a Creative Commons Attribution Non-commercial 4.0 International (CC BY-NC) licence.

    Download (4.57MB)

Check full text

DOI (Published version): 10.1108/hff-10-2024-0800

Abstract

This paper introduces a Particle Trickle Release (PTR) algorithm for implementing an inlet velocity boundary condition in Graph Network Simulators (GNS) and explores the ability of GNS to extrapolate and apply the learned fluid dynamics to unseen, out-of-distribution examples.The study uses the ...

Full description

Published in: International Journal of Numerical Methods for Heat & Fluid Flow
ISSN: 0961-5539 1758-6585
Published: Emerald Publishing Limited 2025
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa69333
first_indexed 2025-04-23T22:22:43Z
last_indexed 2025-11-12T08:12:09Z
id cronfa69333
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2025-11-11T11:54:42.8829169</datestamp><bib-version>v2</bib-version><id>69333</id><entry>2025-04-23</entry><title>Graph network simulators (GNS) for modelling particle-based fluid flow with a given inlet velocity</title><swanseaauthors><author><sid>748e66fe0d153ead026b97c26fcaecc5</sid><firstname>Philip</firstname><surname>Pe</surname><name>Philip Pe</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>0136f9a20abec3819b54088d9647c39f</sid><ORCID>0000-0003-4848-4545</ORCID><firstname>Rajesh</firstname><surname>Ransing</surname><name>Rajesh Ransing</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2025-04-23</date><abstract>This paper introduces a Particle Trickle Release (PTR) algorithm for implementing an inlet velocity boundary condition in Graph Network Simulators (GNS) and explores the ability of GNS to extrapolate and apply the learned fluid dynamics to unseen, out-of-distribution examples.The study uses the 'WaterRamps' training dataset, which provides essential parameters for fluid particles. The training of the GNS is conducted using both the existing dynamics bootstrapping method and a sequential training approach to assess their effectiveness in capturing fluid dynamics accurately. The PTR algorithm is introduced to ensure realistic particle inflows at boundaries, calculated using a binomial distribution based on inflow velocity and inlet boundary length.The PTR algorithm demonstrated realistic particle release with minimal errors in particle count and area consistency compared to theoretical values. Sequential training resulted in a mean squared error (MSE) of 13.9 &#xD7; 10-3, slightly higher than the 12.9 &#xD7; 10-3 achieved with dynamics bootstrapping. The study also highlights challenges in maintaining incompressibility conditions and the tendency to learn excessive wall friction, which leads to undesired boundary layer development, particularly in out-of-distribution simulations such as the 'WaterVortex' example and flow over a backward-facing step.This paper contributes to the field of graph network-based fluid flow modelling by facilitating the implementation of inlet velocity conditions through the PTR algorithm and evaluating the effectiveness of sequential training. The degree of compressibility is assessed using a newly proposed velocity divergence term, and a &#x2018;push particle&#x2019; algorithm is introduced to improve the quality of particle distribution.</abstract><type>Journal Article</type><journal>International Journal of Numerical Methods for Heat &amp; Fluid Flow</journal><volume>35</volume><journalNumber>9</journalNumber><paginationStart>3053</paginationStart><paginationEnd>3079</paginationEnd><publisher>Emerald Publishing Limited</publisher><placeOfPublication/><isbnPrint/><isbnElectronic/><issnPrint>0961-5539</issnPrint><issnElectronic>1758-6585</issnElectronic><keywords>Deep learning, Data-driven modelling, Physics-informed modelling, Particle-based fluid flow, Lagrangian fluid flow, Graph neural networks</keywords><publishedDay>27</publishedDay><publishedMonth>10</publishedMonth><publishedYear>2025</publishedYear><publishedDate>2025-10-27</publishedDate><doi>10.1108/hff-10-2024-0800</doi><url/><notes/><college>COLLEGE NANME</college><CollegeCode>COLLEGE CODE</CollegeCode><institution>Swansea University</institution><apcterm>Not Required</apcterm><funders>The authors gratefully acknowledge the support of the Supercomputing Wales project, which is part-funded by the European Regional Development Fund (ERDF) through the Welsh Government.</funders><projectreference/><lastEdited>2025-11-11T11:54:42.8829169</lastEdited><Created>2025-04-23T23:13:58.1741967</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering</level></path><authors><author><firstname>Philip</firstname><surname>Pe</surname><order>1</order></author><author><firstname>Rajesh</firstname><surname>Ransing</surname><orcid>0000-0003-4848-4545</orcid><order>2</order></author></authors><documents><document><filename>69333__34075__4375a0bd65174cd2b47eae6496b5a591.pdf</filename><originalFilename>particle_trickle_algorithm - AAM.pdf</originalFilename><uploaded>2025-04-23T23:21:36.4052875</uploaded><type>Output</type><contentLength>4789510</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><documentNotes>This author accepted manuscript is deposited under a Creative Commons Attribution Non-commercial 4.0 International (CC BY-NC) licence.</documentNotes><copyrightCorrect>true</copyrightCorrect><language>eng</language><licence>https://creativecommons.org/licenses/by-nc/4.0/deed.en</licence></document></documents><OutputDurs/></rfc1807>
spelling 2025-11-11T11:54:42.8829169 v2 69333 2025-04-23 Graph network simulators (GNS) for modelling particle-based fluid flow with a given inlet velocity 748e66fe0d153ead026b97c26fcaecc5 Philip Pe Philip Pe true false 0136f9a20abec3819b54088d9647c39f 0000-0003-4848-4545 Rajesh Ransing Rajesh Ransing true false 2025-04-23 This paper introduces a Particle Trickle Release (PTR) algorithm for implementing an inlet velocity boundary condition in Graph Network Simulators (GNS) and explores the ability of GNS to extrapolate and apply the learned fluid dynamics to unseen, out-of-distribution examples.The study uses the 'WaterRamps' training dataset, which provides essential parameters for fluid particles. The training of the GNS is conducted using both the existing dynamics bootstrapping method and a sequential training approach to assess their effectiveness in capturing fluid dynamics accurately. The PTR algorithm is introduced to ensure realistic particle inflows at boundaries, calculated using a binomial distribution based on inflow velocity and inlet boundary length.The PTR algorithm demonstrated realistic particle release with minimal errors in particle count and area consistency compared to theoretical values. Sequential training resulted in a mean squared error (MSE) of 13.9 × 10-3, slightly higher than the 12.9 × 10-3 achieved with dynamics bootstrapping. The study also highlights challenges in maintaining incompressibility conditions and the tendency to learn excessive wall friction, which leads to undesired boundary layer development, particularly in out-of-distribution simulations such as the 'WaterVortex' example and flow over a backward-facing step.This paper contributes to the field of graph network-based fluid flow modelling by facilitating the implementation of inlet velocity conditions through the PTR algorithm and evaluating the effectiveness of sequential training. The degree of compressibility is assessed using a newly proposed velocity divergence term, and a ‘push particle’ algorithm is introduced to improve the quality of particle distribution. Journal Article International Journal of Numerical Methods for Heat & Fluid Flow 35 9 3053 3079 Emerald Publishing Limited 0961-5539 1758-6585 Deep learning, Data-driven modelling, Physics-informed modelling, Particle-based fluid flow, Lagrangian fluid flow, Graph neural networks 27 10 2025 2025-10-27 10.1108/hff-10-2024-0800 COLLEGE NANME COLLEGE CODE Swansea University Not Required The authors gratefully acknowledge the support of the Supercomputing Wales project, which is part-funded by the European Regional Development Fund (ERDF) through the Welsh Government. 2025-11-11T11:54:42.8829169 2025-04-23T23:13:58.1741967 Faculty of Science and Engineering School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering Philip Pe 1 Rajesh Ransing 0000-0003-4848-4545 2 69333__34075__4375a0bd65174cd2b47eae6496b5a591.pdf particle_trickle_algorithm - AAM.pdf 2025-04-23T23:21:36.4052875 Output 4789510 application/pdf Accepted Manuscript true This author accepted manuscript is deposited under a Creative Commons Attribution Non-commercial 4.0 International (CC BY-NC) licence. true eng https://creativecommons.org/licenses/by-nc/4.0/deed.en
title Graph network simulators (GNS) for modelling particle-based fluid flow with a given inlet velocity
spellingShingle Graph network simulators (GNS) for modelling particle-based fluid flow with a given inlet velocity
Philip Pe
Rajesh Ransing
title_short Graph network simulators (GNS) for modelling particle-based fluid flow with a given inlet velocity
title_full Graph network simulators (GNS) for modelling particle-based fluid flow with a given inlet velocity
title_fullStr Graph network simulators (GNS) for modelling particle-based fluid flow with a given inlet velocity
title_full_unstemmed Graph network simulators (GNS) for modelling particle-based fluid flow with a given inlet velocity
title_sort Graph network simulators (GNS) for modelling particle-based fluid flow with a given inlet velocity
author_id_str_mv 748e66fe0d153ead026b97c26fcaecc5
0136f9a20abec3819b54088d9647c39f
author_id_fullname_str_mv 748e66fe0d153ead026b97c26fcaecc5_***_Philip Pe
0136f9a20abec3819b54088d9647c39f_***_Rajesh Ransing
author Philip Pe
Rajesh Ransing
author2 Philip Pe
Rajesh Ransing
format Journal article
container_title International Journal of Numerical Methods for Heat & Fluid Flow
container_volume 35
container_issue 9
container_start_page 3053
publishDate 2025
institution Swansea University
issn 0961-5539
1758-6585
doi_str_mv 10.1108/hff-10-2024-0800
publisher Emerald Publishing Limited
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 - Mechanical Engineering{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Aerospace, Civil, Electrical, General and Mechanical Engineering - Mechanical Engineering
document_store_str 1
active_str 0
description This paper introduces a Particle Trickle Release (PTR) algorithm for implementing an inlet velocity boundary condition in Graph Network Simulators (GNS) and explores the ability of GNS to extrapolate and apply the learned fluid dynamics to unseen, out-of-distribution examples.The study uses the 'WaterRamps' training dataset, which provides essential parameters for fluid particles. The training of the GNS is conducted using both the existing dynamics bootstrapping method and a sequential training approach to assess their effectiveness in capturing fluid dynamics accurately. The PTR algorithm is introduced to ensure realistic particle inflows at boundaries, calculated using a binomial distribution based on inflow velocity and inlet boundary length.The PTR algorithm demonstrated realistic particle release with minimal errors in particle count and area consistency compared to theoretical values. Sequential training resulted in a mean squared error (MSE) of 13.9 × 10-3, slightly higher than the 12.9 × 10-3 achieved with dynamics bootstrapping. The study also highlights challenges in maintaining incompressibility conditions and the tendency to learn excessive wall friction, which leads to undesired boundary layer development, particularly in out-of-distribution simulations such as the 'WaterVortex' example and flow over a backward-facing step.This paper contributes to the field of graph network-based fluid flow modelling by facilitating the implementation of inlet velocity conditions through the PTR algorithm and evaluating the effectiveness of sequential training. The degree of compressibility is assessed using a newly proposed velocity divergence term, and a ‘push particle’ algorithm is introduced to improve the quality of particle distribution.
published_date 2025-10-27T06:46:41Z
_version_ 1851283997900931072
score 11.090362

Similar Items