No Cover Image

Journal article 420 views 123 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
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 × 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.
Keywords: Deep learning, Data-driven modelling, Physics-informed modelling, Particle-based fluid flow, Lagrangian fluid flow, Graph neural networks
College: Faculty of Science and Engineering
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.
Issue: 9
Start Page: 3053
End Page: 3079

Similar Items