Machine Learning Model for Gas–Liquid Interface Reconstruction in CFD Numerical Simulations
The volume of fluid (VoF) method is widely used in multiphase flow simulations to track and locate the interface between two immiscible fluids. The relative volume fraction in each cell is used to recover the interface properties (i.e., normal, location, and curvature). Accurate computation of the l...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Fluids |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2311-5521/10/1/20 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588486304071680 |
|---|---|
| author | Tamon Nakano Michele Alessandro Bucci Jean-Marc Gratien Thibault Faney |
| author_facet | Tamon Nakano Michele Alessandro Bucci Jean-Marc Gratien Thibault Faney |
| author_sort | Tamon Nakano |
| collection | DOAJ |
| description | The volume of fluid (VoF) method is widely used in multiphase flow simulations to track and locate the interface between two immiscible fluids. The relative volume fraction in each cell is used to recover the interface properties (i.e., normal, location, and curvature). Accurate computation of the local interface curvature is essential for evaluation of the surface tension force at the interface. However, this interface reconstruction step is a major bottleneck of the VoF method due to its high computational cost and low accuracy on unstructured grids. Recent attempts to apply data-driven approaches to this problem have outperformed conventional methods in many test cases. However, these machine learning-based methods are restricted to computations on structured grids. In this work, we propose a machine learning-enhanced VoF method based on graph neural networks (GNNs) to accelerate interface reconstruction on general unstructured meshes. We first develop a methodology for generating a synthetic dataset based on paraboloid surfaces discretized on unstructured meshes to obtain a dataset akin to the configurations encountered in industrial settings. We then train an optimized GNN architecture on this dataset. Our approach is validated using analytical solutions and comparisons with conventional methods in the OpenFOAM framework on a canonical test. We present promising results for the efficiency of GNN-based approaches for interface reconstruction in multiphase flow simulations in the industrial context. |
| format | Article |
| id | doaj-art-726861e1765b4108a45e73a6260595c9 |
| institution | Kabale University |
| issn | 2311-5521 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Fluids |
| spelling | doaj-art-726861e1765b4108a45e73a6260595c92025-01-24T13:32:37ZengMDPI AGFluids2311-55212025-01-011012010.3390/fluids10010020Machine Learning Model for Gas–Liquid Interface Reconstruction in CFD Numerical SimulationsTamon Nakano0Michele Alessandro Bucci1Jean-Marc Gratien2Thibault Faney3Institut National de Recherche en Informatique et en Automatique (INRIA), LISN, bât 660, Université Paris-Saclay, 91405 Orsay, FranceInstitut National de Recherche en Informatique et en Automatique (INRIA), LISN, bât 660, Université Paris-Saclay, 91405 Orsay, FranceIFP Énergies-Nouvelles, 1 et 4 av Bois Préau, 92852 Rueil-Malmaison, FranceIFP Énergies-Nouvelles, 1 et 4 av Bois Préau, 92852 Rueil-Malmaison, FranceThe volume of fluid (VoF) method is widely used in multiphase flow simulations to track and locate the interface between two immiscible fluids. The relative volume fraction in each cell is used to recover the interface properties (i.e., normal, location, and curvature). Accurate computation of the local interface curvature is essential for evaluation of the surface tension force at the interface. However, this interface reconstruction step is a major bottleneck of the VoF method due to its high computational cost and low accuracy on unstructured grids. Recent attempts to apply data-driven approaches to this problem have outperformed conventional methods in many test cases. However, these machine learning-based methods are restricted to computations on structured grids. In this work, we propose a machine learning-enhanced VoF method based on graph neural networks (GNNs) to accelerate interface reconstruction on general unstructured meshes. We first develop a methodology for generating a synthetic dataset based on paraboloid surfaces discretized on unstructured meshes to obtain a dataset akin to the configurations encountered in industrial settings. We then train an optimized GNN architecture on this dataset. Our approach is validated using analytical solutions and comparisons with conventional methods in the OpenFOAM framework on a canonical test. We present promising results for the efficiency of GNN-based approaches for interface reconstruction in multiphase flow simulations in the industrial context.https://www.mdpi.com/2311-5521/10/1/20graph neural networkinterface reconstructionVoF methodunstructured mesh |
| spellingShingle | Tamon Nakano Michele Alessandro Bucci Jean-Marc Gratien Thibault Faney Machine Learning Model for Gas–Liquid Interface Reconstruction in CFD Numerical Simulations Fluids graph neural network interface reconstruction VoF method unstructured mesh |
| title | Machine Learning Model for Gas–Liquid Interface Reconstruction in CFD Numerical Simulations |
| title_full | Machine Learning Model for Gas–Liquid Interface Reconstruction in CFD Numerical Simulations |
| title_fullStr | Machine Learning Model for Gas–Liquid Interface Reconstruction in CFD Numerical Simulations |
| title_full_unstemmed | Machine Learning Model for Gas–Liquid Interface Reconstruction in CFD Numerical Simulations |
| title_short | Machine Learning Model for Gas–Liquid Interface Reconstruction in CFD Numerical Simulations |
| title_sort | machine learning model for gas liquid interface reconstruction in cfd numerical simulations |
| topic | graph neural network interface reconstruction VoF method unstructured mesh |
| url | https://www.mdpi.com/2311-5521/10/1/20 |
| work_keys_str_mv | AT tamonnakano machinelearningmodelforgasliquidinterfacereconstructionincfdnumericalsimulations AT michelealessandrobucci machinelearningmodelforgasliquidinterfacereconstructionincfdnumericalsimulations AT jeanmarcgratien machinelearningmodelforgasliquidinterfacereconstructionincfdnumericalsimulations AT thibaultfaney machinelearningmodelforgasliquidinterfacereconstructionincfdnumericalsimulations |