No description available.

In this dissertation, an experimental and numerical study of dense coarse solids-liquid flows has been performed. The experimental work mainly involved pressure drop measurements in a vertical flow loop. A limited number of measurements of solids velocity profiles were also obtained in the upward flow section of the flow loop. The numerical work involved simulations of coarse particles-in-water flows in vertical and horizontal pipes. The vertical flow simulations were performed using the commercial CFD software, ANSYS CFX-4.4, while ANSYS CFX-10 was used to simulate the flows in the horizontal pipes. The simulations were performed to investigate the applicability of current physically-based models to very dense coarse-particle flows.<p> In the experimental study, measurements of pressure drop and local solids velocity profiles were obtained. The experiments were conducted in a 53 mm diameter vertical flow loop using glass beads of 0.5 mm and 2.0 mm diameter solids for concentration up to 45%. The liquid phase was water. The measured pressure drop exhibited the expected dependence on bulk velocity and solids mean concentration. The wall shear stress was determined by subtracting the gravitational contribution from the measured pressure drop. For flow with the 0.5 mm particles at high bulk velocities, the values of the wall shear stress were essentially similar for each concentration in the upward flow sections but more variation, indicating the effect of concentration, was noted in the downward flow section. At lower bulk velocities, the wall shear stresses with the 0.5 mm glass beads-water flow showed a dependence on concentration in both test sections. This was attributed to an increase in the slip velocity. For the large particle (2.0 mm glass beads), similar observations were made but the effect of concentration was much less in the upward test section. In the downward test section, the wall shear stress for the flow of the 2.0 mm glass beads increased by almost a constant value for the bulk velocities investigated. The solids velocity profiles showed that the solids velocity gradient is large close to the wall. In addition, the solids velocity profiles indicated that the slip velocity increased at lower velocities due to increase in the bulk concentration in the upward flow section.<p> For the vertical flow simulations, different physical models based on the kinetic theory of granular flows were programmed and implemented in ANSYS CFX-4.4. These models, referred to as the kf-ef-ks-es, kf-ef-ks-es-Ts and kf-ef-ks-kfs models, were investigated by focusing on the closure laws for the solids-phase stress. The treatment of the granular temperature Ts depends on whether small- or large-scale fluctuating motion of the particles is considered. The models were implemented via user-Fortran routines. The predicted results were compared with available experimental results. The predicted solids-phase velocity profiles matched the measured data quite well close to the pipe wall but over-predicted it in the core region. The solids concentration, on the other hand, was significantly under-predicted for concentrations higher than 10%. Variations in the predictions of the phasic turbulent kinetic energy and the eddy viscosity were noted; the effect of solids concentration on them was mixed. A general conclusion drawn from the work is that a more accurate model is required for accurate and consistent prediction of coarse particle flows at high concentrations (less than 10%). In a related study, attention was given to wall boundary conditions again focusing on the effect of the solids-phase models at the wall. Comparison between numerical predictions, using some of the existing wall boundary condition models for the solids phase in particulate flows, with experimental results indicated that the physical understanding of the influence of the fluid and solids-phase on each other and their effect on frictional head loss is far from complete. The models investigated failed to reproduce the experimental results. At high solids concentration, it was apparent from the present study that the no-slip and free-slip wall boundary conditions are not appropriate for liquid-solid flows.<p> For the horizontal flow case, three-dimensional simulations were performed with a focus on the velocity and concentration distributions. Medium and coarse sand-in-water flows in three pipe diameters were considered to investigate the default solids stress models in ANSYS CFX-10. Simulations were performed for three cases by considering: 1) no additional solids-phase stress, i.e. no model for Ts; 2) a zero equation, and 3) an algebraic equilibrium model for the granular temperature. The model predictions were compared to experimental results. The effect of particle size, solids-phase concentration, and pipe diameter was explored using the algebraic equilibrium model. All the cases for the models considered exhibited the characteristic features of horizontal coarse particle slurry flows. The zero equation and the algebraic equilibrium model for the granular temperature produced similar results that were not significantly different from the prediction obtained when no solids-phase stress was considered. The comparison with experimental results was mixed. Locally, the measured solids-phase velocity distributions were over-predicted, whereas the solids concentration was reasonably reproduced in the core of all the pipes. The concentration at the bottom and top walls were over-, and under-predicted, respectively. This was attributed to the inappropriate phasic wall boundary condition models available.

During the production and transportation of oil and gas, flow assurance issues may occur due to the solid deposits that are formed and carried by the flowing fluid. Solid deposition may cause serious damage and possible failure to production equipment in the flow lines. The major flow assurance problems that are faced in the fields are concerned with asphaltene, wax and scale deposition, as well as hydrate formations. Hydrates, wax and asphaltene deposition are mostly addressed in deep-water environments, where fluid flows through a long path with a wide range of pressure and temperature variations (Hydrates are generated at high pressure and low temperature conditions). In fact, a large change in the thermodynamic condition of the fluid yields phase instability and triggers solid deposit formations. In contrast, scales are formed in aqueous phase when some incompatible ions are mixed. Among the different flow assurance issues in hydrocarbon reservoirs, asphaltenes are the most complicated one. In fact, the difference in the nature of these molecules with respect to other hydrocarbon components makes this distinction. Asphaltene molecules are the heaviest and the most polar compounds in the crude oils, being insoluble in light n-alkenes and readily soluble in aromatic solvents. Asphaltene is attached to similarly structured molecules, resins, to become stable in the crude oils. Changing the crude oil composition and increasing the light component fractions destabilize asphaltene molecules. For instance, in some field situations, CO₂ flooding for the purpose of enhanced oil recovery destabilizes asphaltene. Other potential parameters that promote asphaltene precipitation in the crude oil streams are significant pressure and temperature variation. In fact, in such situations the entrainment of solid particulates in the flowing fluid and deposition on different zones of the flow line yields serious operational challenges and an overall decrease in production efficiency. The loss of productivity leads to a large number of costly remediation work during a well life cycle. In some cases up to $5 Million per year is the estimated cost of removing the blockage plus the production losses during downtimes. Furthermore, some of the oil and gas fields may be left abandoned prematurely, because of the significance of the damage which may cause loss about $100 Million. In this dissertation, we developed a robust wellbore model which is coupled to our in-house developed compositional reservoir model (UTCOMP). The coupled wellbore/reservoir simulator can address flow restrictions in the wellbore as well as the near-wellbore area. This simulator can be a tool not only to diagnose the potential flow assurance problems in the developments of new fields, but also as a tool to study and design an optimum solution for the reservoir development with different types of flow assurance problems. In addition, the predictive capability of this simulator can prescribe a production schedule for the wells that can never survive from flow assurance problems. In our wellbore simulator, different numerical methods such as, semi-implicit, nearly implicit, and fully implicit schemes along with blackoil and Equation-of-State compositional models are considered. The Equation-of-State is used as state relations for updating the properties and the equilibrium calculation among all the phases (oil, gas, wax, asphaltene). To handle the aqueous phase reaction for possible scales formation in the wellbore a geochemical software package (PHREEQC) is coupled to our simulator as well. The governing equations for the wellbore/reservoir model comprise mass conservation of each phase and each component, momentum conservation of liquid, and gas phase, energy conservation of mixture of fluids and fugacity equations between three phases and wax or asphaltene. The governing equations are solved using finite difference discretization methods. Our simulation results show that scale deposition is mostly initiated from the bottom of the wellbore and near-wellbore where it can extend to the upper part of the well, asphaltene deposition can start in the middle of the well and the wax deposition begins in the colder part of the well near the wellhead. In addition, our simulation studies show that asphaltene deposition is significantly affected by CO₂ and the location of deposition is changed to the lower part of the well in the presence of CO₂. Finally, we applied the developed model for the mechanical remediation and prevention procedures and our simulation results reveal that there is a possibility to reduce the asphaltene deposition in the wellbore by adjusting the well operation condition. / text

La compréhension des écoulements à bulles dans les centrales nucléaires demeure encore un élément limitant dans l’analyse des opérations et de la sûreté des installations. Pour ne citer qu’un exemple, l’amélioration de la durée de vie etde la performance des générateurs de vapeur nécessite d’appréhender les régimes d’écoulement au sein des tubes qui sont responsables de leur vibration. Cependant, pour simuler avec précision ces écoulements, les codes de simulation numérique doivent relever de nombreux défis parmi lesquels la capacité à simuler des inclusions ayant des tailles très variées. Dans cette thèse, une nouvelle approche, appelée approche multi-champ, est implémentée dans le code NEPTUNE_CFD, basé sur un modèle bi-fluide. Cette approche inclut une méthode de suivi d’interface pour les grandes structures déformables et prend en compte les effets liés à la turbulence et aux changements de phase.Pour simuler de tels écoulements complexes en limitant le coût CPU, l’approche multi-champ considère séparément les petites inclusions sphériques des grandes inclusions déformables. Ainsi, les petites structures sphériques sont définies via un champ eulérien dispersé évoluant au sein d’un champ continu porteur, comme c’est habituellement le cas avec le modèle bi-fluide. Les grosses bulles déformables sont considérées comme des interfaces entre deux champs continus, un champ liquide et un champ gaz. Si on prend l’exemple d’un écoulement diphasique avec de l’eau et des bulles d’air de différentes tailles, trois champs sont alors définis pour cet écoulement: un champ continu liquide, un champ continu gaz et un champ dispersé gaz contenant les petites bulles sphériques. Cependant, simuler avec précision des interfaces entre deux champscontinus avec le modèle bi-fluide nécessite le développement de traitements spécifiques afin de coupler les deux champs à l’interface et de limiter la diffusion de cette interface.Après avoir amélioré la simulation des interfaces dans des écoulements laminaires, les effets liés à la turbulence sont étudiés. Une étude a priori de simulations aux grandes échelles est proposée pour identifier les termes sous-mailles et comparer différents modèles de turbulence disponibles dans la littérature. L’implémentation et la validation du modèle de turbulence retenu suite à l’étude sont détaillées. Les changements de phase sont ensuite explorés via le développement d’un modèle spécifique pour le terme de transfert de masse. Pour finir, des simulations trois champs sont présentées. De nouveauxcritères sont définis pour modéliser la fragmentation des grandes inclusions déformables en petites bulles sphériques ainsi que la coalescence de ces dernières pour former de grandes bulles déformables.A chaque étape de l’implémentation des différents modèles évoqués, des validations basées sur des données analytiques et issues d’expériences sont présentées afin de s’assurer que les phénomènes physiques sont bien prédits. Des cas tests dans des configurations industrielles sont également détaillés pour montrer la capacité de l’approche développée à simuler des écoulements complexes / Bubbly flows occurring in nuclear power plants remain a major limiting phenomenon for the analysis of operation and safety. As an example, the improvement of steam generator lifetime and performance relies on the comprehension of flow regimes inside the tubes responsible for tube vibrations. However, to ensure an accurate simulation of these flows, theComputational Multi-Fluid Dynamics (CMFD) codes have to take up many challenges, among others the ability of dealing with a variety of inclusion sizes. The classical two-fluid model allows simulating small spherical inclusions but is not able to compute large deformable inclusions. Thus, in this thesis, a new approach, called the multifield approach, is implementedin the CMFD code NEPTUNE_CFD, based on a two-fluid model. This approach includes an interface tracking method for large and deformable structures and takes into account turbulence and phase change effects.To simulate such complex flows with reasonable computational costs, the multifield approach considers separately the small spherical inclusions and the large deformable ones. Thus, the small spherical structures are defined as a dispersed field evolving in a continuous carrier field, as usually done in the two-fluid model. The large deformable bubbles are considered as interfaces between two continuous phases treated as two different fields in the two-fluid model. In the example of a two-phase flow with water and air bubbles of different sizes, three fields are defined: a continuous liquid field, a continuous gas field and a dispersed gas field containing the small spherical bubbles. However, the accurate simulation of interfaces between the two continuous fields within the two-fluid model requires specific treatments to couple the two fields at the interface and to limit the interface smearing.After improving the interface simulation in laminar flows, turbulence effects are investigated. An a priori Large Eddy Simulation (LES) study is performed to identify the predominant subgrid terms and to compare different availableturbulence models. The implementation and validation of the most suitable model is proposed. Phase change interfaces are then explored with the development of a specific model for the mass transfer term. Finally, three fields simulations are performed. New criteria are defined for the breakup of the large deformable inclusions into small spherical bubbles and for the coalescence of the latter forming large deformable bubbles.Validation at each step of the models implementations are presented using analytical and experimental data to ensure that the physical phenomena are well predicted. Test cases in industrial configurations are finally performed to show the ability of the developed approach to deal with complex flows

[pt] O escoamento bifásico no regime de golfadas é um padrão de escoamento que requer esforço em sua caracterização e modelagem, devido às características marcantes da distribuição espacial das fases, que gera intermitência ao escoamento. Este escoamento ocorre na presença de uma grande faixa de vazões de gás e líquido, em tubulações de diâmetro médio e pequeno, com variação periódica da densidade, fração de vazio e pressões na seção transversal da tubulação. No presente trabalho estuda-se numericamente o regime de golfadas ao longo de tubulações horizontais utilizando-se o Modelo de Dois Fluidos. Uma análise detalhada das estatísticas do escoamento é realizada com a caracterização dos principais parâmetros de uma golfada, como comprimento, velocidade da frente e cauda e freqüência de passagem das golfadas ao longo da tubulação. Os dados obtidos das simulações são analisados através de histogramas de distribuição assim como de funções densidade de probabilidade (PDF) das variáveis hidrodinâmicas, destacando-se o caráter estocástico do escoamento no padrão de golfadas. Adicionalmente, comparações com dados de trabalhos experimentais da literatura foram realizadas com concordância muito boa. / [en] Slug flow is a two-phase flow pattern that requires large effort in its characterization and modeling, due to special characteristics of the phase’s spatial distribution, which causes flow intermittency. This flow occurs in a wide range of gas and liquid flow rates in pipes of medium and small diameters, with periodic variation of density, void fraction and pressures in pipe-cross-section. This work presents a numerical study of the slug regime through horizontal pipes using the Two-Fluid Model. A detailed statistical analysis of the flow was carried out with characterization of main slug parameters, such as slug length, front and tail velocities and slug frequency along the pipeline. The numerical results were analyzed through distribution histograms as well as probability density function (PDF) of the hydrodynamic variables, showing the stochastic characteristic of slug flow pattern. Further, comparisons with experimental data from the literature were performed, showing very good agreement.

Due to the high-speed flow in a chute spillway, cavitation damages often occur. This undesired phenomenon threatens the safety of the structure. For the purpose of eliminating the damages, an aerator is often installed in the spillway. To understand its characteristics, physical model tests are a popular method. To complement the model tests, computation fluid dynamics (CFD) simulations are used to study aerator flows. To represent the two-phase flows, multiphase models should be employed. This thesis examines two of them, namely, the Volume-Of-Fluid model (VOF) and Two-Fluid model. Based on the background of the Bergeforsen dam, the aerator flow is modelled by means of the VOF model. The simulated spillway discharge capacity is in accordance with the experimental data. Compared with the results, empirical formulas fail to evaluate the air supply capacity of aerator as it is wider than the conventional width. A hypothetical vent modification is proposed. For the original and proposed layouts, the study illustrates the difference in the air-flow conditions. The results show that a larger vent area is, for a large-width aerator, preferable in the middle of the chute. To study the flip bucket-shaped aerators in the Gallejaur dam, physical model tests and prototype observations are conducted. The results lead to contradicting conclusions in terms of jet breakup and air entrainment. A CFD model is, as an option, employed to explain the reason of the discrepancy. The numerical results coincide with the prototype observations. The jet breakup and air entrainment are evaluated from air cavity profiles; the air-pressure drops are small in the cavity. The discrepancy is due to overestimation of the surface-tension effect in the physical model tests. Based on the experimental data of an aerator rig at the Laboratory of Hydraulics, Hydrology and Glaciology (VAW), ETH Zurich, the Two-Fluid model is used to predict air concentration distributions in the aerated flow. The model includes relevant forces governing the motion of bubbles and considers the effects of air bubble size. The numerical results are conformable to the experiments in the air cavity zone. Downstream of the cavity, the air concentration near the chute bottom is higher, which is presumably caused by the fact that the interfacial forces in the Two-Fluid model are underestimated. / <p>QC 20170224</p>

[pt] Uma análise numérica de escoamento anular vertical ascendente é realizada utilizando o Modelo de Dois Fluidos unidimensional. Para escoamentos verticais, na ausência de mecanismos estabilizadores, o sistema de equações resultante é incondicionalmente mal posto. Dessa forma, visando tornar o sistema de equações bem posto, adicionou-se às equações de quantidade de movimento um termo de pressão dinâmica. Dois modelos disponíveis na literatura são investigados. O primeiro só considera a pressão dinâmica na equação de quantidade de movimento do líquido, a qual é baseada em uma simples expressão para a velocidade da onda na interface. O segundo modelo inclui uma contribuição da pressão dinâmica em ambas equações de quantidade de movimento. No presente trabalho, um terceiro modelo é proposto, o qual é baseado no primeiro modelo, com uma avaliação mais realista da velocidade da onda na interface. O efeito estabilizante da pressão dinâmica é demonstrado através de um rigoroso teste de convergência de malha. As equações de conservação são discretizadas com o método de volumes finitos, com uma integração temporal de primeira ordem e uma discretização espacial TVD de segunda ordem. Tanto o segundo quanto o terceiro modelo considerados apresentaram solução independente da malha. Parâmetros do escoamento como gradiente de pressão, espessura do filme, e variáveis características da onda obtidos numericamente são comparados com dados experimentais disponíveis na literatura, apresentando boa concordância. / [en] A numerical analysis of vertical ascending annular flow with the 1D Two-Fluid model is performed. It is well known that, in vertical flows, the resulting system of equations is unconditionally ill-posed in the absence of stabilizing mechanisms. Therefore, in the present work, to render the system of equations well-posed, modeling of dynamic pressure is included in the momentum equations. Two models available in the literature are examined. The first one only considers the dynamic pressure in the liquid momentum equation, which is based on a simple expression for the interface wave velocity. The second model includes a dynamic pressure contribution to both momentum equations. A third model is proposed based on the first, with a more realistic estimation of the interface wave velocity. A systematic grid convergence test is performed to demonstrate the stabilizing effect of the dynamic pressure. The conservation equations are discretized with the finite volume method, with a first order time integration, and a second order TVD spatial discretization. A grid-independent solution can be found when applying the second and third models considered. Flow parameters such as pressure drop, film thickness and wave characteristics numerically obtained are compared against available experimental data, presenting good agreement.

Motivées par des données expérimentales sur la magnétisation de réseau de nanofils de plomb, les résolutions numériques des équations stationnaires de Ginzburg-Landau (GL) se sont focalisées sur les géométries à symétrie axiale. L'effet Meissner, les états représentant un vortex d'Abrikosov ou encore des Vortex Géants (``GiantVortex') centrés à l'origine du cylindre ont alors pu être identifiés sous l’hypothèse d’invariance sous rotation selon l’axe de symétrie du cylindre étudié (modèle à une dimension, 1D). En identifiant le type de transition par le caractère continu ou non du paramètre d'ordre autour du changement de phase, une frontière à l'échelle mésoscopique a également pu être identifiée au travers du modèle 1D. Plus spécifiquement, la limite entre les deux types de transitions décrite par le paramètre phénoménologique κ = λ /ξ ( =1/√2 à l’échelle macroscopique) devient une fonction non constante dépendant à la fois du rayon normalisé, u=R/λ, et de la vorticité L: κ =f(u,L). Les deux longueurs caractéristiques λ et ξ représentent respectivement les longueurs de pénétration et de cohérence d’un échantillon supraconducteur. Une comparaison avec les résultats obtenus par Zharkov permet de valider notre démarche numérique employée pour la résolution numérique des équations de GL à une dimension. En employant un modèle à deux dimensions (2D), la symétrie sous rotation des solutions a également été relâchée. Basée sur le principe de moindre action, la résolution propose alors un schéma numérique indépendant du type d'équations du mouvement à solutionner. Les configurations du type MultiVortex ont alors pu être identifiées, et comparées aux solutions du groupe du Professeur F. Peeters. Ces différents accords ont confirmé la démarche développée. Une modélisation de la magnétisation expérimentale d'un réseau de nanofils a également été développée. De par la taille réduite des nanofils, l'interaction magnétique entre ceux-ci a pu être négligée. La magnétisation totale du réseau est alors construite par une sommation incluant la contribution individuelle en magnétisation de chaque fil, pondérée par un poids reflétant une distribution gaussienne pour les rayons des fils constituant le réseau. La magnétisation individuelle est évidemment obtenue par résolution des équations du mouvement de GL précédemment étudiées avec les modèles 1D et 2D. En ajustant les paramètres libres associés à ce modèle décrivant la magnétisation totale du réseau, les données expérimentales ont pu être reproduites endéans 10% de marge d'erreur, l'intervalle d'incertitude caractéristique de la théorie effective de Ginzburg-Landau. Ces variables attachées au modèle de la magnétisation totale, reprennent la valeur moyenne m et l'écart-type s de la distribution gaussienne, ainsi que les longueurs caractéristiques λ(T) et ξ(T) présentes dans la théorie de GL. Un test totalement indépendant de l'analyse des magnétisations a permis de valider les valeurs déterminées pour la distribution des rayons. Les grandeurs ajustées pour les longueurs λ(T) et ξ(T) ont fait l'objet d'une analyse supplémentaire en termes de leur dépendance en température et du libre parcours moyen des électrons. Malgré l'accord entre les données expérimentales et la magnétisation théorique, il est important de mentionner qu'un paramètre libre supplémentaire, associé à l'apparition de configurations décrivant un vortex magnétique, a dû être introduit. Il modifie empiriquement la métastabilité trop longue en mode champ externe décroissant de l'état décrivant un vortex d'Abrikosov. La correction expulse donc le vortex avant sa prédiction théorique liée à la disparition de la barrière de Bean-Linvingston. Une étude plus approfondie de cette barrière de potentiel fut donc également réalisée. Cependant, elle n'est pas concluante en regard des données expérimentales analysées. Il n'en demeure pas moins que la transition apparaît dans un domaine en champ magnétique cohérent vis-à-vis de la description en énergie libre des états de vorticités voisines d'une unité de quantum de flux magnétique. La correspondance entre les longueurs caractéristiques du modèle phénoménologique de GL et les longueurs issues des théories microscopiques de Pippard et BCS a également abordée. Cette étude permet entre autre de comparer les différentes dépendances possibles en température avec les longueurs obtenues de l'analyse de magnétisation des nanofils en plomb. Au delà de l'accord avec le modèle des deux-fluides de Gorter et Casimir, une extrapolation bien en deçà de la température critique Tc est proposée pour les paramètres phénoménologiques λ(T) et ξ(T) de Ginzburg-Landau. Même si la correspondance entre les magnétisations expérimentales et théoriques semblait déjà l'indiquer, il est possible d'appliquer les équations de Ginzburg-Landau pour décrire le comportement magnétique du plomb bien en deçà de sa température critique. De plus, les paramètres associés possèdent une dépendance tout à fait conforme à une autre théorie empirique, le modèle des deux-fluides. Basée sur le modèle de Pippard, une détermination de la valeur du libre parcours moyen des normaux a également été isolée. Elle justifie alors une distinction entre les deux échantillons analysés en terme de leur degré d'impureté. Les résultats électrons obtenus étant en accord avec les procédures de fabrication des nanofils de plomb, cette nouvelle constatation, positive avec l'expérience, confirme une fois de plus la cohérence du modèle développé pour la magnétisation totale, et justifie l'emploi des équations de GL à toutes les températures en dessous de Tc. / Mesoscopic superconductors are described within the framework of the nonlinear Ginzburg-Landau theory. The two coupled nonlinear equations are solved numerically and we investigate the properties, in particular the order of the transition and the vortex configurations, of cylinders submitted to an external magnetic field. Meissner state, Abrikosov vortices, GiantVortex and MultiVortex solutions are described. The Bean-Livingston barrier in mesoscopic cylinders is also numerically studied. This theoretical work was applied to understand experimental magnetizations of lead nanowires in an array well below the superconducting transition temperature Tc. By freely adjusting the GL phenomenological lengths λ (T) and ξ (T), the experimental magnetization curves are reproduced to within a 10% error margin. The Meissner and the Abrikosov state were also experimentally observed in this apparently type-I superconductor. This fact is a consequence of the non-trivial behaviour of the critical boundary κ _c ($=1/√2 in bulk materials) between type-I and type-II phase transition at mesoscopic scales. Beyond the experimental-theoretical agreement, the question whether the GL model remains valid far below Tc is also addressed. The temperature dependence of the adjusted characteristic lengths is compared with different theoretical and empirical laws. The best agreement is achieved for the Gorter-Casimir two-fluid model. A comparison between lead nanowire arrays electrodeposited under constant and pulsed voltage conditions allows us to distinguish both samples in terms of their electronic mean free paths. The characterisation of the latter quantities concurs perfectly with the experimental expectation given the different electrodeposition techniques.

Cette thèse contribue à la vérification et à la validation du modèle bifluide de Baer-Nunziato, pour modéliser les phénomènes de transitoires hydrauliques dans les réseaux de tuyauteries industrielles. Il s’agit d’abord de modéliser les écoulements de transitoires hydrauliques avec le modèle bifluide en représentation eulérienne, puis d’étendre ce modèle en formalisme ALE (Arbitrary Lagrangian Eulerian) pour prendre en compte l’interaction fluide-structure (IFS). Pour modéliser les écoulements, des lois de fermetures du modèle bifluide concernant les termes interfaciaux, les termes sources et les lois thermodynamiques (EOS) ont d’abord été étudiées. Ensuite, le système complet a été simulé avec une méthode à pas fractionnaires qui admet deux étapes, l’une pour la résolution de la partie convective, l’autre pour les termes sources. L’ensemble de schémas a été vérifié et étendu aux EOS ‘Stiffened Gas généralisées’ afin de représenter le changement de phase eau-vapeur. Après avoir retrouvé certains phénomènes typiques associés aux transitoires hydrauliques, le modèle bifluide a été validé avec l’expérience de Simpson, l’expérience Canon, et comparé avec deux modèles homogènes sur ces deux expériences. Enfin, une version ALE du modèle bifluide a été mise en œuvre et vérifiée sur un cas de propagation d’ondes de pression dans une conduite flexible. La variation de la célérité des ondes dans le fluide liée au couplage fluide/structure a été bien retrouvée. La validation a été effectuée sur un cas expérimental d'explosion dans une tuyauterie en eau. Les simulations sont en bon accord avec les données expérimentales. / This thesis contributes to the verification and the validation of the Baer-Nunziato (BN) model, to modelize water hammer phenomena in industrial piping systems. It consists of two parts, the first is to modelize water hammer flows with the BN model in Eulerian representation and the second is to extend this model to the ALE (Arbitrary Lagrangian Eulerian) formalism so as to take into account fluid-structure interaction (FSI). To modelize water hammer flows, closure laws of the BN model concerning the interfacial/source terms and the equations of state (EOS) were first studied. Then the whole system was simulated with a fractional step method including two steps, one for the resolution of the convective part, the other for the source terms. All schemes have been extended to ‘generalized Stiffened Gas’ EOS in order to represent phase-change. After regaining some typical phenomena associated with water hammer flows, the BN model was validated with the Simpson experiment, a classical water hammer test case, and the Canon experience, a rapid decompression of fluid in a high pressure duct. Moreover, the model was compared with two homogeneous models on both experiments. Finally, an ALE version of the BN model was implemented, and verified on a case of wave propagation in a ‘single’ phase flow and a two-phase flow in a flexible pipe. The variation of wave propagation speed due to the coupling between the fluid and the structure has been well retrieved. The validation was performed on an experiment which examines the response of a pipe filled with water, subjected to a violent pressure peak (140 bar). The simulations are in good agreement with the experimental data.

Reatores de leito fluidizado circulante são intensamente utilizados em aplicações de larga escala como craqueamento catalítico de petróleo e combustão de carvão. Projeto e desenvolvimento nestas áreas são fortemente baseados em plantas de demonstração, a custos elevadíssimos. Nesse contexto, tratamentos utilizando mecânica dos fluidos computacional assumem considerável relevância. Os complexos padrões de escoamento gás-sólido que se desenvolvem nos reatores de leito fluidizado circulante determinam taxas de reação e exigem, portanto, descrições hidrodinâmicas rigorosas. Os modelos Eulerianos do contínuo ou de dois fluidos são correntemente considerados a escolha mais prática na busca destas descrições. Estas formulações são baseadas na aplicação da hipótese de meio contínuo tanto para fases macroscopicamente contínuas quanto dispersas. Neste trabalho, desenvolve-se simulações numéricas para descrever processos hidrodinâmicos e reativos em reatores de leito fluidizado circulante aplicando modelagem de dois fluidos. Considera-se situações típicas de combustão em leito fluidizado circulante de carvão mineral. Como processo reativo considera-se a absorção de dióxido de enxofre por calcário. Realiza-se análises hidrodinâmicas, e de efeitos hidrodinâmicos sobre a reação de interesse. Realiza-se simulações de regime permanente e de regime transiente. Mostra-se que as simulações de regime permanente permitem análises qualitativas do processo, e provêem condições iniciais para simulações transientes diretamente no regime de escoamento estatisticamente permanente. Em relação à hidrodinâmica, conclui-se que os modelos de dois fluidos correntes são ainda bastante crus. São claramente necessárias melhores descrições reológicas e relações constitutivas sub-grade mais acuradas. Em relação à reação química, conclui-se que predições reativas acuradas somente poderão ser obtidas se descrições hidrodinâmicas rigorosas forem combinadas com descrições reativas igualmente rigorosas. / Circulating fluidized bed reactors are widely used in large scale applications such as catalytic cracking of petrol and coal combustion. Development and design in those areas are strongly based on demonstration plants, at extremely high costs. In this context, treatments applying computational fluid mechanics assume considerable relevance. The complex gas-solid flow patterns which develop inside the circulating fluidized bed reactors determine reaction rates, so that rigorous hydrodynamic descriptions are required. The continuum Eulerian or two-fluid models are currently considered the more practical choice for providing such descriptions. Those formulations are based on the application of the continuum hypothesis for both macroscopically continuous and dispersed phases. In this work numerical simulation is performed to describe both hydrodynamics and reactive processes in circulating fluidized beds applying two-fluid modeling. Typical situations of circulating fluidized bed coal combustion are considered. The reactive process considered is the absorption of sulfur dioxide by limestone. Analyses are performed of hydrodynamics, and regarding hydrodynamic effects over the concerning reaction. Both steady state and transient simulations are performed. It is shown that steady state simulations allow qualitative analyses, and do provide initial conditions for transient runs straightly inside the statistical steady state flow regime. Concerning hydrodynamics, it is concluded that the current two-fluid models are still very crude. Clearly, better rheological descriptions are required alongside with more accurate sub-grid constitutive relations. Regarding chemical reaction, it is concluded that accurate reactive predictions shall only be found if rigorous hydrodynamic descriptions are combined with equally rigorous reaction descriptions.