Skip to main content

RECONSTRUCCION DE DATOS ENRIQUECIDA POR APRENDIZAJE AUTOMATICO DE FLUJO DE FLUIDOS PARA INTERACCION VIRTUAL CON SISTEMAS DE INGENIERIA

PID2023-150840OA-I00

•
Funding agency name Agencia Estatal de Investigación
Funding agency acronym AEI
Program Programa Estatal para Impulsar la Investigación Científico-Técnica y su Transferencia
Subprogram Subprograma Estatal de Generación de Conocimiento
Call Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Call year 2023
State Plan 2021-2023 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre UNIVERSITAT POLITECNICA DE CATALUNYA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 5
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Physics-driven efficient modeling of pseudo-boiling regions in transcritical wall-bounded flows for thermal management applications

  • Elmansy, Reda Mohamed Yousyf Abdallah|||0000-0002-7484-317X
  • Barea Sánchez, Guillem|||0009-0004-0448-0715
  • Jofre Cruanyes, Lluís|||0000-0003-2437-259X
Dataset link: https://doi.org/10.25452/figshare.plus.29787812. 2026 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/bync-nd/4.0/ )., Accurate prediction of thermal behavior and pseudo-boiling regions in high-pressure transcritical flows is critical for the design of advanced wall-bounded thermal management systems, ranging from aerospace propulsion to high-efficiency power cycles. However, high-fidelity three-dimensional (3D) scale-resolving simulations are often computationally prohibitive for design and optimization. This study introduces a physics-driven compute-efficient framework that leverages a quasi-3D low-fidelity model to accelerate the analysis of such flows. In this approach, the low-fidelity model is first utilized to predict wall temperature profiles and characterize the wall-normal topology of the pseudo-boiling region. These thermal distributions are subsequently used to inform 3D direct numerical simulations (DNS), streamlining the overall design process. The results demonstrate that the methodology accurately reproduces mean thermal distributions and bulk flow quantities across a wide range of operating pressures and wall thermal configurations. Specifically, the error in predicting the pseudo-boiling wall-normal location remains below 5%, while trends in the bulk Reynolds number are consistently captured. Although the low-fidelity model does not resolve shear-dependent turbulence statistics, it accurately predicts the position of the pseudo-boiling region and bulk flow parameters. This framework provides a robust, cost-effective tool for the preliminary design and optimization of high-pressure thermal systems, significantly reducing the reliance on conventional, resource-intensive DNS., This work is funded by the European Union (ERC, SCRAMBLE, 101040379). Views and opinions expressed are however those of the authors only and do not necessarily reflect those of the European Union or the European Research Council. Neither the European Union nor the granting authority can be held responsible for them. The authors gratefully acknowledge the SGR program (2021-SGR-01045) of Generalitat de Catalunya (Catalonia), and the PID2023-150840OA-I00 grant of Agencia Estatal de Investigación (AEI, Spain)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Thermal effects on high-pressure transcritical wall-bounded turbulence

  • Elmansy, Reda Mohamed Yousyf Abdallah|||0000-0002-7484-317X
  • Barea Sánchez, Guillem|||0009-0004-0448-0715
  • Mellibovsky Elstein, Fernando|||0000-0003-0497-9052
  • Jofre Cruanyes, Lluís|||0000-0003-2437-259X
© 2025 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC) license (https://creativecommons.org/licenses/by-nc/4.0/), The sustainment of turbulence under high-pressure transcritical conditions is examined using direct numerical simulations of channel flow subject to various combinations of fluid pressure and wall temperatures. The working fluid is carbon dioxide at supercritical reduced pressures P=Pc∈{1.5, 2, 5}P=Pc∈{1.5,2,5} (the subscript cc denotes the critical value) driven through a channel bounded by differentially heated isothermal walls at reduced temperatures Tcw=Tc∈{0.8, 0.9, 0.95}Tcw=Tc∈{0.8,0.9,0.95} (cold wall) and Thw=Tc∈{1.1, 1.2, 1.4}Thw=Tc∈{1.1,1.2,1.4} (hot wall). The resulting friction Reynolds numbers fall within the ranges Rescw∈[56, 111]Rescw∈[56,111] at the cold wall and Reshw∈[118, 167]Reshw∈[118,167] at the hot wall. The computational domain has dimensions (Lx,Ly,Lz)=(2π, 2, 2π/3)d(Lx,Ly,Lz)=(2π,2,2π/3)d in the streamwise, wall-normal, and spanwise directions, corresponding to a minimal flow unit (dd is the channel half-height). The results show that setting the temperature of the hot wall slightly above the pseudo-boiling point, specifically in the range Thw/Tpb∈[1.01, 1.18]Thw/Tpb∈[1.01,1.18], maximizes the Nusselt number associated with both the hot and cold walls, while the bulk and friction Reynolds numbers peak in the narrower range Thw/Tpb∈[1.01, 1.10]Thw/Tpb∈[1.01,1.10]. Furthermore, both turbulent activity and heat transfer across the channel are found to be highest in the early supercritical regime, with pressures just above critical. These findings highlight the importance of tuning the thermophysical parameters precisely if turbulence is to be sustained and heat transfer enhanced in high-pressure transcritical fluid flows., This work was funded by the European Union (ERC, SCRAMBLE, No. 101040379). Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union or the European Research Council. Neither the European Union nor the granting authority can be held responsible for them. The authors acknowledge the support from the SGR program (No. 2021-SGR-01045) of the Generalitat de Catalunya (Catalonia) and the No. PID2023-150840OA-I00 grant of the Agencia Estatal de Investigación (AEI, Spain)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Eigenspace-based reinforcement learning acceleration of statistical convergence in turbulent flows

  • Masclans Serrat, Núria
  • Jofre Cruanyes, Lluís|||0000-0003-2437-259X
Direct numerical simulation approaches play a critical role in advancing the understanding and accurate modeling of turbulent flow phenomena. However, the direct simulation of high-Reynolds-number turbulent flows is computationally intensive, requiring high spatial resolution and extended simulation times to achieve statistically reliable results. The strategy leverages deep reinforcement learning to apply dynamic, data-driven perturbations to the Reynolds stresses eigenspace. This allows an optimized control policy to guide the evolution of the Reynolds stresses tensor across all degrees of freedom (magnitude, anisotropy and orientation) within a physics-constrained framework. For the present numerical experiments, a reduced three-dimensional action space targeting magnitude and shape perturbations is utilized as an initial proof-of-concept. To that end, the framework is thoroughly described and its performance evaluated in the context of canonical turbulent channel flows at friction Reynolds numbers 𝑅𝑒𝜏 = 100 and 𝑅𝑒𝜏 = 180. The trained agent successfully accelerates the convergence of key flow statistics, including the fluctuating velocity components and the underlying turbulent structure, i.e., anisotropy, particularly in the 𝑅𝑒𝜏 = 100 training case. A comprehensive computational cost analysis demonstrates that the training overhead is rapidly amortized during unsupervised evaluation phases.Depending on the training duration and early-stopping criteria, the initial training overhead is recovered within 3 to 10 deployment runs under the executed conservative time steps, and within 2 to 5 runs under the projected CFL-optimal conditions. This yields a net computational saving when accelerating the convergence of the fluctuating velocity components to their targeted error tolerances when extrapolating from the 𝑅𝑒𝜏 = 100 training case to the 𝑅𝑒𝜏 = 180 deployment case. Furthermore, the evaluation provided critical insights into the complex nature of the control problem, identifying main performance trade-offs during training, the strong dependency on the agent’s actuation frequency, and the fundamental challenges of policy generalization to a new flow regime. This work validates, therefore, the presented approach as a viable tool for targeting the convergence acceleration of specific flow statistics, and demonstrates a clear path toward more computationally efficient high-fidelity simulations of turbulence., The authors gratefully acknowledge the Formació de Professorat Universitari scholarship (FPU-UPC 2022) of Universitat Politècnica de Catalunya (UPC), the SGR program (2021-SGR-01045) of Generalitat de Catalunya (Catalonia), and the PID2023-150840OA-I00 grant of Agencia Estatal de Investigación (AEI, Spain).
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Thermal effects on high-pressure transcritical wall-bounded turbulence

  • El Mansy, Reda
  • Barea Sánchez, Guillem
  • Mellibovsky Elstein, Fernando
  • Jofre Cruanyes, Lluís
© 2025 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC) license (https://creativecommons.org/licenses/by-nc/4.0/), The sustainment of turbulence under high-pressure transcritical conditions is examined using direct numerical simulations of channel flow subject to various combinations of fluid pressure and wall temperatures. The working fluid is carbon dioxide at supercritical reduced pressures P=Pc∈{1.5, 2, 5}P=Pc∈{1.5,2,5} (the subscript cc denotes the critical value) driven through a channel bounded by differentially heated isothermal walls at reduced temperatures Tcw=Tc∈{0.8, 0.9, 0.95}Tcw=Tc∈{0.8,0.9,0.95} (cold wall) and Thw=Tc∈{1.1, 1.2, 1.4}Thw=Tc∈{1.1,1.2,1.4} (hot wall). The resulting friction Reynolds numbers fall within the ranges Rescw∈[56, 111]Rescw∈[56,111] at the cold wall and Reshw∈[118, 167]Reshw∈[118,167] at the hot wall. The computational domain has dimensions (Lx,Ly,Lz)=(2π, 2, 2π/3)d(Lx,Ly,Lz)=(2π,2,2π/3)d in the streamwise, wall-normal, and spanwise directions, corresponding to a minimal flow unit (dd is the channel half-height). The results show that setting the temperature of the hot wall slightly above the pseudo-boiling point, specifically in the range Thw/Tpb∈[1.01, 1.18]Thw/Tpb∈[1.01,1.18], maximizes the Nusselt number associated with both the hot and cold walls, while the bulk and friction Reynolds numbers peak in the narrower range Thw/Tpb∈[1.01, 1.10]Thw/Tpb∈[1.01,1.10]. Furthermore, both turbulent activity and heat transfer across the channel are found to be highest in the early supercritical regime, with pressures just above critical. These findings highlight the importance of tuning the thermophysical parameters precisely if turbulence is to be sustained and heat transfer enhanced in high-pressure transcritical fluid flows., This work was funded by the European Union (ERC, SCRAMBLE, No. 101040379). Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union or the European Research Council. Neither the European Union nor the granting authority can be held responsible for them. The authors acknowledge the support from the SGR program (No. 2021-SGR-01045) of the Generalitat de Catalunya (Catalonia) and the No. PID2023-150840OA-I00 grant of the Agencia Estatal de Investigación (AEI, Spain)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Minimal flow unit of wall-bounded high-pressure transcritical turbulence

  • Elmansy, Reda Mohamed Yousyf Abdallah
  • Monteiro, Carlos Alexandre Bandarrinha
  • Mellibovsky Elstein, Fernando
  • Jofre Cruanyes, Lluís
The minimum domain size required to maintain fully developed wall-bounded turbulent flow in high-pressure transcritical regimes is analyzed using direct numerical simulations. The computations involve carbon dioxide at supercritical conditions with pressure P=Pc ¼ 2 (subscript c denotes critical value), confined between cold (temperature T=Tc ¼ 0:85) and hot (T=Tc ¼ 1:5) isothermal walls. The corresponding friction Reynolds numbers are Res 76 and 123 for the cold and hot walls, respectively. The study considers a large baseline domain of size Lx=d ¼ 4p, Ly=d ¼ 2, and Lz=d ¼ð8=3Þp in the streamwise, wall-normal, and spanwise directions, respectively, with d the half-channel height, as well as truncated setups along the streamwise and/or spanwise directions. The results show that rather small domains are sufficient to reproduce near-wall flow motions, while larger domains are needed to properly capture large-scale structures, especially within the log-law region. Additionally, the study confirms that the distance over which thermal interactions occur is shorter than the typical size of hydrodynamic structures, particularly so in the spanwise direction. This finding suggests that the minimum periodic domain size required in the high-pressure transcritical regime is comparable to that of the minimum flow unit established by other researchers for wall-bounded flows under standard pressure-temperature conditions., This work is funded by the European Union (ERC, SCRAMBLE, 101040379). Views and opinions expressed are, however, those of the authors only and do not necessarily reflect those of the European Union or the European Research Council. Neither the European Union nor the granting authority can be held responsible for them. The authors also acknowledge support from the SGR program (2021-SGR-01045) of the Generalitat de Catalunya (Spain), the PID2023-150840OA-I00 grant of the Agencia Estatal de Investigación (AEI, Spain), and the computer resources at FinisTerrae III & MareNostrum and the technical support provided by CESGA & Barcelona Supercomputing Center (RES-IM-2023-1-0005, RES-IM-2023-2-0005). F.M. acknowledges support from the PID2020-114043GB-I00 and PID2023-150029NB-I00 grants of the AEI., Peer Reviewed, Postprint (published version)