1,720,981 research outputs found
Numerical prediction of turbulent flow and heat transfer in buoyancy-affected liquid metal flows
Get PDFThe present paper investigates the capabilities of some selected Reynolds-Averaged Navier Stokes (RANS) based turbulence models in reproducing liquid metals thermal hydraulics Direct Numerical Simulation (DNS) data. For this purpose, forced and mixed convection conditions, both addressing buoyancy-aided and buoyancy-opposed flow, are considered. The paper mainly focuses on velocity and temperature fields estimation, providing a comparison between the RANS and DNS computations. The capabilities of the turbulence models are discussed with the aim to highlight which ones provide the best predictions. In particular, attention is paid to the approach adopted for the calculation of the turbulent heat flux contributions. Together with models assuming the commonly adopted Simple Gradient Diffusion Hypothesis (SGDH) approach and the Reynolds analogy, a model including the Algebraic Heat Flux Model (AHFM) approach is considered. While being a practical and robust approach to deal with turbulent heat fluxes, the SGDH approach shows intrinsic limitation in dealing with liquid metal thermal hydraulics, mainly because of their low-Prandtl number. The adoption of a more advanced AHFM method may instead relevantly improve the quality of the obtained predictions. The obtained results show that the selected model adopting the AHFM method provides definitively better predictions of the addressed phenomena with respect to the ones considering the SGDH approach. While some discrepancies are still observed for the velocity fields, the temperature fields are captured very well, suggesting a clear superiority of the AHFM model. The present paper thus provides further validation and supports the use of AHFM as a valuable tool to predict turbulent heat fluxes. © 2023 Elsevier Lt
Numerical simulation of mixing buoyant jets: Preliminary studies
No full textPreliminary numerical analyses are reported for the case of three vertical planar mixing jets at different temperatures, in view of a forthcoming reference DNS to be performed in the frame of the SESAME European project. The reference case stems from the well-known PLAJEST triple jet experiment, but with a relative increase of the buoyancy effect, achieved by reducing the Reynolds number by a factor 5. Pre-production DNS runs for the reference case and a complementary RANS parametric analysis with varying Prandtl and Richardson numbers are carried out, highlighting that a suitable configuration for a benchmark can be obtained with a hotter central jet and colder lateral jets
Algebraic heat flux modeling for numerical prediction of heat transfer and flow in Natural Convection
No full textIn nuclear industry applications, from design and safety aspects, it is important to predict the flow and heat transfer. The passive cooling systems, one of the robust cooling systems in the nuclear design are based on natural convections. In the RANS approach, non-linear unknown heat flux term has to be closed by appropriate model for accurate predictions of flow and thermal fields. This thesis presents a numerical framework for simulating heat and flow transfer in natural convection flow regimes by employing algebraic turbulent heat flux model AHFM-NRG+, coupled with second order turbulence model. The turbulent heat flux model coefficients are calibrated and a correlation between Rayleigh and Prandtl numbers with model coefficient is defined. The model is employed for different test cases of Rayleigh Bénard convections and is validated by comparing the simulation results with reference DNS data.Aerospace Engineerin
Assessment of the NEK5000 code for Direct Numerical Simulations
No full textThe present work is included in the frame of the fundamental studies performed worldwide aiming at improving the prediction tools for investigating turbulence phenomena, in order to provide the basis for the assessment of codes and models suitable for the detailed analysis of complex fluid systems, like nuclear reactors.
A critical aspect of the nuclear reactor safety is the fluid-structure interaction, and in particular, the combined effect of a turbulent flow with a thermal mixing, which may lead to local Thermal Fatigue phenomena into the structures due to Thermal Striping.
The main goal of the present work is to assess the DNS capabilities of the spectral elements code NEK5000 for the fundamental turbulent flows and their thermal interaction with solid structures. This work is divided into two parts.In the first part, DNS computations are carried out in order to evaluate the performances of the code in predicting the turbulent flow behaviour for simplistic channel flow configurations.
Whereas, in the second part, the DNS analyses are further extended for a conjugate heat transfer case, i.e. a planer channel flow along with heated walls. Three different temperature fields are considered.
A wide range of numerical parameters are tested and their influence is studied in order to obtain high quality turbulence statistics both for the velocity and the thermal fields. The obtained results are compared against other well-known DNS databases available in literature. The comparison suggests that NEK5000 exhibits excellent capabilities to perform high quality DNS computations
Numerical Prediction of Natural Circulation Heat Transfer for Supercritical Carbon Dioxide
No full textDue to their high specific heat, low viscosity, and good diffusivity, supercritical fluids have the potential to be ideal coolants. However, understanding the heat transfer for fluids under supercritical conditions has been a challenge. To understand the peculiar heat transfer characteristics, a wide range of experiments with different ranges of parameters and geometrical configurations has been conducted. The generated experimental data can be used as a reference to expand and assess the prediction accuracy of computational fluid dynamics models under supercritical conditions. Out of these models, RANS is the most widely used and consumes less computational power relative to other models. In this paper, natural circulation heat transfer of supercritical carbon dioxide will be investigated using RANS approach. To validate the prediction accuracy of RANS model, an extensive comparative study with experimental data is presented in the present paper
Contribution à la simulation numérique des décollements d'écoulements turbulents induits par choc. Application à l'écoulement sur-détenu de tuyère supersonique
No full textShock-induced flow separation and reattachment are encountered in many configurations, such as supersonic inlets, transonic airfoils or rocket nozzles. These phenomena involve complex interactions of boundary layers with compression or expansion waves and exhibit a low-frequency unsteady behaviour which still requires a clear explanation. This study aims at better identifying the physical mechanisms which drive the global structure of these flows and suggesting improved numerical tools in order to predict these more accurately. The appearance of free and restricted separations in supersonic annular jets occuring in thrust optimised contour nozzles operating in overexpanded conditions is more particularly investigated while various hypothesis are tested to explain the evolution of the associated unsteady asymmetric wall pressure field in function of the nozzle pressure ratio. The numerical strategy proposed relies on a realizable extension of the Detached Eddy Simulation, combined with high order shock capturing schemes and an implicit time integration algorithm. This methodology is applied for a wide range of both constant or transient inflow conditions and leads to identify more accurately the appearance of free and restricted separations and the time-varying morphology of the flow during the transition process. For both regimes, the simulation is carried out for long-enough time to perform reliable statistical analysis and azimuthal expansion of the wall pressure field and thus investigate extensively the possible origins of the side-load activities.Les décollements d'écoulement induits par choc et leur éventuel réattachement sur paroi sont observés dans de nombreuses configurations d'intérêt pratique, incluant les entrées d'air, les profils transsoniques ou les tuy`eres de lanceurs spatiaux. Ces phénom`enes mettent en jeu des interactions complexes entre couches limites et ondes de choc ou de détente conduisant à des instationnarités à basses fréquences dont l'origine reste aujourd'hui à élucider. Cette étude vise d'une part à proposer une stratégie numérique permettant de prévoir plus précisément ces phénom`enes de décollement et d'autre part d'identifier les principaux mécanismes physiques qui pilotent l'évolution de leur structure globale. L'étude porte plus particuli`erement sur les configurations de décollements libres ou séparés apparaissant en tuy`ere optimisée en poussée opérant en régime surdétendu. Différents mod`eles phénoménologiques sont ainsi testées pour décrire l'évolution du champ de pression instationnaire et dissymétrique en fonction du niveau de surdétente. La stratégie numérique proposée repose sur la combinaison de schémas à capture de choc d'ordre élevé (WENO 5), d'algorithmes d'intégration implicite en temps et d'une modélisation de la turbulence étendant l'approche Detached Eddy Simulation via l'ajout de corrections de réalisabilité. Une large plage de niveaux de surdétente est considérée, à la fois en condition d'entrée stabilisée et transitoire, afin de clarifier les conditions d'existence des différents régimes de décollements libres et restreints, ainsi que l'évolution temporelle de la morphologie globale de l'écoulement transitant entre ces deux régimes. L'évolution instationnaire de l'écoulement est simulée sur des temps suffisamment longs pour permettre une analyse spectrale des contributions des premiers modes azimutaux à la dynamique basse fréquence du champ de pression pariétale
Contribution à la simulation numérique des décollements d'écoulements turbulents induits par choc. Application à l'écoulement sur-détenu de tuyère supersonique
No full textShock-induced flow separation and reattachment are encountered in many configurations, such as supersonic inlets, transonic airfoils or rocket nozzles. These phenomena involve complex interactions of boundary layers with compression or expansion waves and exhibit a low-frequency unsteady behaviour which still requires a clear explanation. This study aims at better identifying the physical mechanisms which drive the global structure of these flows and suggesting improved numerical tools in order to predict these more accurately. The appearance of free and restricted separations in supersonic annular jets occuring in thrust optimised contour nozzles operating in overexpanded conditions is more particularly investigated while various hypothesis are tested to explain the evolution of the associated unsteady asymmetric wall pressure field in function of the nozzle pressure ratio. The numerical strategy proposed relies on a realizable extension of the Detached Eddy Simulation, combined with high order shock capturing schemes and an implicit time integration algorithm. This methodology is applied for a wide range of both constant or transient inflow conditions and leads to identify more accurately the appearance of free and restricted separations and the time-varying morphology of the flow during the transition process. For both regimes, the simulation is carried out for long-enough time to perform reliable statistical analysis and azimuthal expansion of the wall pressure field and thus investigate extensively the possible origins of the side-load activities.Les décollements d'écoulement induits par choc et leur éventuel réattachement sur paroi sont observés dans de nombreuses configurations d'intérêt pratique, incluant les entrées d'air, les profils transsoniques ou les tuy`eres de lanceurs spatiaux. Ces phénom`enes mettent en jeu des interactions complexes entre couches limites et ondes de choc ou de détente conduisant à des instationnarités à basses fréquences dont l'origine reste aujourd'hui à élucider. Cette étude vise d'une part à proposer une stratégie numérique permettant de prévoir plus précisément ces phénom`enes de décollement et d'autre part d'identifier les principaux mécanismes physiques qui pilotent l'évolution de leur structure globale. L'étude porte plus particuli`erement sur les configurations de décollements libres ou séparés apparaissant en tuy`ere optimisée en poussée opérant en régime surdétendu. Différents mod`eles phénoménologiques sont ainsi testées pour décrire l'évolution du champ de pression instationnaire et dissymétrique en fonction du niveau de surdétente. La stratégie numérique proposée repose sur la combinaison de schémas à capture de choc d'ordre élevé (WENO 5), d'algorithmes d'intégration implicite en temps et d'une modélisation de la turbulence étendant l'approche Detached Eddy Simulation via l'ajout de corrections de réalisabilité. Une large plage de niveaux de surdétente est considérée, à la fois en condition d'entrée stabilisée et transitoire, afin de clarifier les conditions d'existence des différents régimes de décollements libres et restreints, ainsi que l'évolution temporelle de la morphologie globale de l'écoulement transitant entre ces deux régimes. L'évolution instationnaire de l'écoulement est simulée sur des temps suffisamment longs pour permettre une analyse spectrale des contributions des premiers modes azimutaux à la dynamique basse fréquence du champ de pression pariétale
A-priori estimation of turbulent length scales for DNS and LES mesh guidelines; Application to Backward facing Step
No full textLa fluidodinamica computazionale (CFD) riveste un ruolo chiave nella descrizione del moto di flussi turbolenti. Delle tre principali tipologie di simulazioni associate in CFD alla turbolenza, Direct Numerical Simulation (DNS), Large Eddy Simulation (LES) e Reynolds Averaged Navier Stokes Equations (RANS), simulazioni altamente affidabili di tipo DNS e LES sono sempre più richieste, in particolare nell'ambito della sicurezza industriale e nucleare.
Nonostante questo, a causa delle fitte mesh necessarie alla descrizione del dominio di calcolo DNS e LES, l'uso di questi metodi risulta particolarmente oneroso in termini di tempo, potenza e costi computazionali. Le grandi risorse richieste da DNS e LES rendono cruciale una corretta generazione della mesh, per la quale, al presente stato dell'arte, non esiste un vero e proprio processo formale. Il presente lavoro di tesi, svolto in Olanda presso l'azienda NRG dell'ambito dei servizi nucleari, mira a proporre un criterio per la generazione della mesh di DNS e LES, al fine di ottimizzare la potenza di calcolo impiegata.
Il metodo proposto si incentra sull'analisi delle scale di lunghezza della turbolenza, computate utilizzando più economiche simulazioni di tipo RANS. Il criterio è dapprima testato su una configurazione Channel Flow, attraverso l'utilizzo di due diversi modelli di simulazioni RANS, ciascuno applicato a fluidi aventi tre diversi numeri di Reynolds. Infine, il metodo è valutato a posteriori su una configurazione più complessa, ossia Backward-Facing Step (BFS) flow. Su questa si sono sviluppate una simulazione di tipo RANS e una di tipo LES. La mesh della LES è stata concepita secondo i metodi convenzionali e, anche attraverso il confronto con dati sperimentali reperiti in letteratura, si paragonano i risultati ottenuti con quelli raggiungibili applicando il procedimento proposto.
In particolare, si è considerata l'abilità del metodo di individuare i punti critici della griglia LES nonché di prevedere adeguati valori delle dimensioni di griglia.
Computational Fluid Dynamics (CFD) plays a key role in the description of turbulent flows, through three main models: Direct Numerical Simulation (DNS), Large Eddy Simulation (LES) e Reynolds Averaged Navier Stokes Equations (RANS). In the industrial and nuclear safety field, DNS and LES are the most requested models, since they are the most reliable and accurate of the three. However, because of the fine mesh they require to describe the domain, DNS and LES are extremely time consuming and costly in terms of computational power. At the present state-of-art, no formal method exists for generating correct mesh for DNS and LES. The present thesis work was developed during an internship at NRG, a Dutch nuclear services provider. This paper proposes a criterion for generating DNS and LES mesh to optimize the computational power requested. The proposed method is based on the turbulence length scales evaluation, through the more lightweight RANS simulations. Firstly, the criterion is tested in the Channel Flow configuration, where two different RANS models are applied to three fluids with different Reynolds Number each. Then, an a-posteriori evaluation is conducted in a more complex configuration, the Backward-Facing Step (BFS) flow. A RANS simulation and a LES, with an empirically set mesh, are computed and their results are compared with each other and with experimental data found in the literature. The mesh suggested by the method is compared both with the actual mesh used in LES and with the match between LES and experimental data. Lastly, the method's capability in predicting both the correct grid sizes modifications and the location of the discrepancies between LES and experimental data is discussed
Contribution à la simulation numérique des décollements d'écoulements turbulents induits par choc. Application à l'écoulement sur-détenu de tuyère supersonique
No full textShock-induced flow separation and reattachment are encountered in many configurations, such as supersonic inlets, transonic airfoils or rocket nozzles. These phenomena involve complex interactions of boundary layers with compression or expansion waves and exhibit a low-frequency unsteady behaviour which still requires a clear explanation. This study aims at better identifying the physical mechanisms which drive the global structure of these flows and suggesting improved numerical tools in order to predict these more accurately. The appearance of free and restricted separations in supersonic annular jets occuring in thrust optimised contour nozzles operating in overexpanded conditions is more particularly investigated while various hypothesis are tested to explain the evolution of the associated unsteady asymmetric wall pressure field in function of the nozzle pressure ratio. The numerical strategy proposed relies on a realizable extension of the Detached Eddy Simulation, combined with high order shock capturing schemes and an implicit time integration algorithm. This methodology is applied for a wide range of both constant or transient inflow conditions and leads to identify more accurately the appearance of free and restricted separations and the time-varying morphology of the flow during the transition process. For both regimes, the simulation is carried out for long-enough time to perform reliable statistical analysis and azimuthal expansion of the wall pressure field and thus investigate extensively the possible origins of the side-load activities.Les décollements d'écoulement induits par choc et leur éventuel réattachement sur paroi sont observés dans de nombreuses configurations d'intérêt pratique, incluant les entrées d'air, les profils transsoniques ou les tuy`eres de lanceurs spatiaux. Ces phénom`enes mettent en jeu des interactions complexes entre couches limites et ondes de choc ou de détente conduisant à des instationnarités à basses fréquences dont l'origine reste aujourd'hui à élucider. Cette étude vise d'une part à proposer une stratégie numérique permettant de prévoir plus précisément ces phénom`enes de décollement et d'autre part d'identifier les principaux mécanismes physiques qui pilotent l'évolution de leur structure globale. L'étude porte plus particuli`erement sur les configurations de décollements libres ou séparés apparaissant en tuy`ere optimisée en poussée opérant en régime surdétendu. Différents mod`eles phénoménologiques sont ainsi testées pour décrire l'évolution du champ de pression instationnaire et dissymétrique en fonction du niveau de surdétente. La stratégie numérique proposée repose sur la combinaison de schémas à capture de choc d'ordre élevé (WENO 5), d'algorithmes d'intégration implicite en temps et d'une modélisation de la turbulence étendant l'approche Detached Eddy Simulation via l'ajout de corrections de réalisabilité. Une large plage de niveaux de surdétente est considérée, à la fois en condition d'entrée stabilisée et transitoire, afin de clarifier les conditions d'existence des différents régimes de décollements libres et restreints, ainsi que l'évolution temporelle de la morphologie globale de l'écoulement transitant entre ces deux régimes. L'évolution instationnaire de l'écoulement est simulée sur des temps suffisamment longs pour permettre une analyse spectrale des contributions des premiers modes azimutaux à la dynamique basse fréquence du champ de pression pariétale
- …
