Dorian Dupuy
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Dorian Dupuy has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- Adrien Toutant
Laboratoire Procédés, Matériaux et Energie Solaire
11 shared publications Françoise Bataille
PROMES CNRS, Université de Perpignan Via Domitia, Rambla de la Thermodynamique, Tecnosud, 66100 Perpignan, France
6 shared publicationsF. Bataille
PROMES CNRS, Université de Perpignan Via Domitia , Rambla de la Thermodynamique, Tecnosud, 66100 Perpignan, France
3 shared publications- Olivier Simonin
Institut de Mécanique des Fluides de Toulouse (IMFT), Université de Toulouse, CNRS, 31400 Toulouse, France
3 shared publications - Renaud Ansart
University of Toulouse, National Polytechnic Institute of Toulouse (Toulouse INP), Centre National de la Recherche Scientifique (CNRS), Laboratoire de Génie Chimique (LGC), 31000 Toulouse, France
3 shared publications - Guillaume Bois
DEN-Service de Thermo-hydraulique et de Mécanique des Fluides (STMF), CEA, Université Paris-Saclay, F-91191 Gif-sur-Yvette, France
2 shared publications
Latest Publications
- Open Access
- 23 September 2026
- Published by Elsevier in Journal of Computational Physics
- Vol. 569
A conservative cut-cell/front-tracking method for interface-resolved two-phase flow simulations
- 23 September 2026
- Published by Elsevier in Journal of Computational Physics
- Vol. 569
Open AccessA conservative cut-cell/front-tracking method for interface-resolved two-phase flow simulations
- 3 April 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Analyse et modélisation de l'interaction entre thermique et turbulence dans les récepteurs solaires à haute température.
Dans les centrales solaires à tour, le flux solaire est concentré vers un récepteur solaire où son énergie est transférée à un fluide caloporteur. L'écoulement au sein du récepteur solaire est turbulent, fortement anisotherme et à bas nombre de Mach. L'optimisation du récepteur solaire exige une meilleure compréhension et modélisation de l'interaction entre la température et la turbulence. Cette thèse cherche à y contribuer selon deux approches. Tout d'abord, on étudie les échanges énergétiques entre les différentes parties de l'énergie totale. On propose pour cela une nouvelle représentation des échanges énergétiques, fondée sur la moyenne de Reynolds. Cette représentation permet la caractérisation, à partir de simulations numériques directes d'un canal plan bipériodique anisotherme, de l'effet du gradient de température sur les échanges énergétiques associées à l'énergie cinétique turbulente dans les domaines spatial et spectral. Ensuite, on étudie la simulation des grandes échelles des équations de bas nombre de Mach. En utilisant les résultats de simulations numériques directes, on identifie les termes sous-mailles spécifiques à modéliser lorsque l'on utilise le filtre classique, non pondéré, et lorsque l'on utilise le filtre de Favre, pondéré par la masse volumique. Dans les deux cas, on évalue a priori la performance de différents modèles sous-mailles. La pertinence des modèles est vérifiée a posteriori par la réalisation de simulation des grandes échelles.
- 3 April 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Analyse et modélisation de l'interaction entre thermique et turbulence dans les récepteurs solaires à haute température.
Dans les centrales solaires à tour, le flux solaire est concentré vers un récepteur solaire où son énergie est transférée à un fluide caloporteur. L'écoulement au sein du récepteur solaire est turbulent, fortement anisotherme et à bas nombre de Mach. L'optimisation du récepteur solaire exige une meilleure compréhension et modélisation de l'interaction entre la température et la turbulence. Cette thèse cherche à y contribuer selon deux approches. Tout d'abord, on étudie les échanges énergétiques entre les différentes parties de l'énergie totale. On propose pour cela une nouvelle représentation des échanges énergétiques, fondée sur la moyenne de Reynolds. Cette représentation permet la caractérisation, à partir de simulations numériques directes d'un canal plan bipériodique anisotherme, de l'effet du gradient de température sur les échanges énergétiques associées à l'énergie cinétique turbulente dans les domaines spatial et spectral. Ensuite, on étudie la simulation des grandes échelles des équations de bas nombre de Mach. En utilisant les résultats de simulations numériques directes, on identifie les termes sous-mailles spécifiques à modéliser lorsque l'on utilise le filtre classique, non pondéré, et lorsque l'on utilise le filtre de Favre, pondéré par la masse volumique. Dans les deux cas, on évalue a priori la performance de différents modèles sous-mailles. La pertinence des modèles est vérifiée a posteriori par la réalisation de simulation des grandes échelles.
- Open Access
- 19 April 2025
- Published by Elsevier in Computers & Fluids
- Vol. 295
Time integration of cut-cell surfaces for geometric conservation, applied to scalar transport with moving interfaces
- 19 April 2025
- Published by Elsevier in Computers & Fluids
- Vol. 295
Open AccessTime integration of cut-cell surfaces for geometric conservation, applied to scalar transport with moving interfaces
- Open Access
- 15 February 2025
- Published by Elsevier in Powder Technology
- Vol. 455
Meso-scale numerical analysis of the role of Van der Waals adhesion and static friction in fluidized beds of fine solids
- 15 February 2025
- Published by Elsevier in Powder Technology
- Vol. 455
Open AccessMeso-scale numerical analysis of the role of Van der Waals adhesion and static friction in fluidized beds of fine solids
- 16 April 2024
- Published by Elsevier in Powder Technology
- Vol. 441
Calibrating the frictional-pressure model from two-fluid simulation of fluidised beds in the defluidisation regime
- 16 April 2024
- Published by Elsevier in Powder Technology
- Vol. 441
Calibrating the frictional-pressure model from two-fluid simulation of fluidised beds in the defluidisation regime
- Open Access
- 10 April 2024
- Published by Cambridge University Press (CUP) in Data-Centric Engineering
- Vol. 5, e10
Using graph neural networks for wall modeling in compressible anisothermal flows
Compressible anisothermal flows, which are commonly found in industrial settings such as combustion chambers and heat exchangers, are characterized by significant variations in density, viscosity, and heat conductivity with temperature. These variations lead to a strong interaction between the temperature and velocity fields that impacts the near-wall profiles of both quantities. Wall-modeled large-eddy simulations (LESs) rely on a wall model to provide a boundary condition, for example, the shear stress and the heat flux that accurately represents this interaction despite the use of coarse cells near the wall, and thereby achieve a good balance between computational cost and accuracy. In this article, the use of graph neural networks for wall modeling in LES is assessed for compressible anisothermal flow. Graph neural networks are a type of machine learning model that can learn from data and operate directly on complex unstructured meshes. Previous work has shown the effectiveness of graph neural network wall modeling for isothermal incompressible flows. This article develops the graph neural network architecture and training to extend their applicability to compressible anisothermal flows. The model is trained and tested a priori using a database of both incompressible isothermal and compressible anisothermal flows. The model is finally tested a posteriori for the wall-modeled LES of a channel flow and a turbine blade, both of which were not seen during training.
- 10 April 2024
- Published by Cambridge University Press (CUP) in Data-Centric Engineering
- Vol. 5, e10
Open AccessUsing graph neural networks for wall modeling in compressible anisothermal flows
Compressible anisothermal flows, which are commonly found in industrial settings such as combustion chambers and heat exchangers, are characterized by significant variations in density, viscosity, and heat conductivity with temperature. These variations lead to a strong interaction between the temperature and velocity fields that impacts the near-wall profiles of both quantities. Wall-modeled large-eddy simulations (LESs) rely on a wall model to provide a boundary condition, for example, the shear stress and the heat flux that accurately represents this interaction despite the use of coarse cells near the wall, and thereby achieve a good balance between computational cost and accuracy. In this article, the use of graph neural networks for wall modeling in LES is assessed for compressible anisothermal flow. Graph neural networks are a type of machine learning model that can learn from data and operate directly on complex unstructured meshes. Previous work has shown the effectiveness of graph neural network wall modeling for isothermal incompressible flows. This article develops the graph neural network architecture and training to extend their applicability to compressible anisothermal flows. The model is trained and tested a priori using a database of both incompressible isothermal and compressible anisothermal flows. The model is finally tested a posteriori for the wall-modeled LES of a channel flow and a turbine blade, both of which were not seen during training.
- 29 February 2024
- Published by Cambridge University Press (CUP) in Journal of Fluid Mechanics
- Vol. 982, A2
Investigation of near-wall particle statistics in CFD-DEM simulations of dense fluidised beds and derivation of an Eulerian particle dynamic wall boundary condition
In two-fluid simulations of gas–solid fluidised beds, the gaseous phase and the particulate phase are modelled as continuous media. The stress exerted by the particulate medium on the container walls should be modelled to predict accurately the bed dynamics. This paper addresses the modelling of sliding particle–wall contacts in two-fluid simulations, based on reference simulations coupling computational fluid dynamics with the discrete element method (CFD-DEM), in which the individual movement of the particles is tracked. The analysis of the CFD-DEM highlights the complex near-wall behaviour of the particles, which is not reproduced by two-fluid models. Nevertheless, the particle–wall shear stress can be expressed based on the total granular pressure within the first cell off the wall. The model is validated for the two-fluid simulation of a bubbling gas–solid fluidised bed of olefin particles in the dense-fluidisation regime.
- 29 February 2024
- Published by Cambridge University Press (CUP) in Journal of Fluid Mechanics
- Vol. 982, A2
Investigation of near-wall particle statistics in CFD-DEM simulations of dense fluidised beds and derivation of an Eulerian particle dynamic wall boundary condition
In two-fluid simulations of gas–solid fluidised beds, the gaseous phase and the particulate phase are modelled as continuous media. The stress exerted by the particulate medium on the container walls should be modelled to predict accurately the bed dynamics. This paper addresses the modelling of sliding particle–wall contacts in two-fluid simulations, based on reference simulations coupling computational fluid dynamics with the discrete element method (CFD-DEM), in which the individual movement of the particles is tracked. The analysis of the CFD-DEM highlights the complex near-wall behaviour of the particles, which is not reproduced by two-fluid models. Nevertheless, the particle–wall shear stress can be expressed based on the total granular pressure within the first cell off the wall. The model is validated for the two-fluid simulation of a bubbling gas–solid fluidised bed of olefin particles in the dense-fluidisation regime.
- 24 April 2023
- Published by Elsevier in Journal of Computational Physics
- Vol. 487
Data-driven wall modeling for turbulent separated flows
- 24 April 2023
- Published by Elsevier in Journal of Computational Physics
- Vol. 487
Data-driven wall modeling for turbulent separated flows
- Open Access
- 9 March 2023
- Published by Cambridge University Press (CUP) in Data-Centric Engineering
- Vol. 4, e7
Modeling the wall shear stress in large-eddy simulation using graph neural networks
As the Reynolds number increases, the large-eddy simulation (LES) of complex flows becomes increasingly intractable because near-wall turbulent structures become increasingly small. Wall modeling reduces the computational requirements of LES by enabling the use of coarser cells at the walls. This paper presents a machine-learning methodology to develop data-driven wall-shear-stress models that can directly operate, a posteriori, on the unstructured grid of the simulation. The model architecture is based on graph neural networks. The model is trained on a database which includes fully developed boundary layers, adverse pressure gradients, separated boundary layers, and laminar–turbulent transition. The relevance of the trained model is verified a posteriori for the simulation of a channel flow, a backward-facing step and a linear blade cascade.
- 9 March 2023
- Published by Cambridge University Press (CUP) in Data-Centric Engineering
- Vol. 4, e7
Open AccessModeling the wall shear stress in large-eddy simulation using graph neural networks
As the Reynolds number increases, the large-eddy simulation (LES) of complex flows becomes increasingly intractable because near-wall turbulent structures become increasingly small. Wall modeling reduces the computational requirements of LES by enabling the use of coarser cells at the walls. This paper presents a machine-learning methodology to develop data-driven wall-shear-stress models that can directly operate, a posteriori, on the unstructured grid of the simulation. The model architecture is based on graph neural networks. The model is trained on a database which includes fully developed boundary layers, adverse pressure gradients, separated boundary layers, and laminar–turbulent transition. The relevance of the trained model is verified a posteriori for the simulation of a channel flow, a backward-facing step and a linear blade cascade.
- 1 February 2021
- Published by Elsevier in International Journal of Heat and Mass Transfer
- Vol. 166
Boundary-condition models of film-cooling holes for large-eddy simulation of turbine vanes
- 1 February 2021
- Published by Elsevier in International Journal of Heat and Mass Transfer
- Vol. 166
Boundary-condition models of film-cooling holes for large-eddy simulation of turbine vanes