This thesis is focused on direct numerical simulation (DNS) of compressible and incompressible fully developed and developing turbulent flows between isothermal walls using a discontinuous Galerkin method (DGM). Three cases (Ma = 0.2, 0.7 and 1.5) of DNS of turbulent channel flows between isothermal walls with Re ~ 2800, based on bulk velocity and half channel width, have been carried out. It is found that a power law seems to scale mean streamwise velocity with Ma slightly better than the more usual log-law. Inner and outer scaling of second-order and higher-order statistics have been analyzed. The linkage between the pressure gradient and vorticity flux on the wall has been theoretically derived and confirmed and they are highly correlated very close to the wall. The correlation coefficients are influenced by Ma, and viscosity when Ma is high. The near-wall spanwise streak spacing increases with Ma. Isosurfaces of the second invariant of the velocity gradient tensor are more sparsely distributed and elongated as Ma increases. DNS of turbulent isothermal-wall bounded flow subjected to favourable and adverse pressure gradient (FPG, APG) at Ma ~ 0.2 and Reref ~ 428000, based on the inlet bulk velocity and the streamwise length of the bottom wall, is also investigated. The FPG/APG is obtained by imposing a concave/convex curvature on the top wall of a plane channel. The flows on the bottom and top walls are tripped turbulent and laminar boundary layers, respectively. It is observed that the first and second order statistics are strongly influenced by the pressure gradients. The cross-correlation coefficients of the pressure gradients and vorticity flux remain constant across the FPG/APG regions of the flat wall. High correlations between the streamwise/wallnormal pressure gradient and the spanwise vorticity are found near the separation region close to the curved top wall. The angle of inclined hairpin structure to streamwise direction of the bottom wall is smaller (flatter) in the FPG region than the APG region. / Thesis (Ph.D, Mechanical and Materials Engineering) -- Queen's University, 2009-12-21 13:59:53.084

[pt] O estudo fundamental de estruturas turbulentas próximas a paredes é de significativa importância devido ao seu papel dominante em inúmeras aplicações de engenharia. Em escoamentos turbulentos, estruturas de diferentes escalas interagem entre si através de um fenômeno complexo caracterizado por um processo contínuo e auto-sustentável. Embora a presença de padrões de movimento coerente pareça ser responsável pela manutenção da turbulência, a evolução e a interação destas estruturas ainda não são completamente entendidas. Para investigar tais fenômenos, foram realizados estudos de escoamentos turbulentos próximos a paredes sobre uma placa plana em um túnel de vento e dentro de um canal de água quadrado. O presente trabalho empregou uma técnica tomográfica de velocimetria por imagem de partícula de alta taxa de medição de campos de velocidade tridimensionais. O processamento da técnica tomo-PIV foi otimizado a fim de melhorar o desempenho do algoritmo e a exatidão dos campos vetoriais medidos. As medições realizadas permitiram a obtenção de estatísticas condicionais das estruturas coerentes típicas da camada limite turbulenta, tais como regiões de baixa e de alta velocidades, varreduras (sweeps), ejeções (ejections) e vórtices. Os resultados quantitativos mostraram vórtices distribuídos por todo o escoamento, entretanto, concentrados na vizinhança das regiões de baixa velocidade e ejeções. Os resultados indicaram que os vórtices estavam indiretamente correlacionados com o processo de produção de turbulência, apesar de serem os principais responsáveis pela dissipação da energia turbulenta. / [en] The fundamental study of turbulent structures near a wall is of significant importance due to its dominant role in innumerous engineering applications. In turbulent flows, eddies of different scales interact with each other in a complex phenomenon of a continuous self-sustaining process. Although the presence of patterns of coherent motion seem to be responsible for the maintenance of turbulence, the evolution and the interaction of these structures are still not completely understood. To investigate such phenomena, near-wall turbulent flows over a smooth at plate in a wind tunnel and inside a square water channel were conducted. The present work employed a high-repetition tomographic PIV technique to measure three dimensional, time-resolved velocity fields. The tomo-PIV processing was optimized in order to improve the algorithm performance and the vector fields accuracy. Conditional statistics of coherent structures, namely, lowand high-speed regions, sweeps, ejections and vortices, were computed. Quantitative results showed vortical structures spread all over the flow, nevertheless, concentrated close to low-speed regions and ejections. Results indicated that vortical structures were indirectly correlated with the process of turbulence production, although they were the main responsible for the turbulent energy dissipation. / [fr] L étude fondamentale des structures turbulentes proche d une paroi est d une significative importance en raison de son rôle dominant dans nombreuses applications d ingênierie. Dans les écoulements turbulents, des structures d échelles différentes interagissent dans un phénomène complexe caractérisé par un processus autonome et continu. Bien que la présence de modéles de mouvements cohérents semble entretenir la turbulence, l évolution et l interaction de ces structures ne sont pas encore complétement compris. Pour analyser ces phénomènes, l étude d écoulements turbulents à proximité de paróis a été effectuées sur une plaque plane dans une soufflerie et à l intérieur d un canal hydraulique de section carrée. Une technique tomographique de vélocimétrie par images de particules pour mesurer de champs de vitesse tridimensionnel a été employée. Le traitement des données a été optimisé afin d améliorer la précision des champs de vecteurs mesurés. Les mesures realisés ont permis d obtenir des statistiques conditionnelles des structures cohérentes, à savoir, les régions à faible et à grande vitesse, les sweeps, les ejections et les tourbillons. Les résultats quantitatives ont montré des structures tourbillonnaires distribués sur tout le ux, cependant, concentrée au voisinage des régions à faible vitesse et des ejections. Les rèsultats ont indiqué que les structures tourbillonnaires étaient indirectement corrélés avec le processus de production de la turbulence, bien qu ils ont été les principaux responsables de la dissipation d énergie turbulente.

Vegetation is commonly modelled as emergent arrays of rigid, circular cylinders. However, the drag coefficient (CD) of real stems or trunks is closer to that of cylinders with a square cross-section. In this thesis, vegetation has been idealised as square cylinders in laboratory experiments with a turbulence intensity of the order of 10% which is similar to that of typical river flows. These cylinders may also represent other obstacles such as architectural structures. This research has determined CD of an isolated cylinder and cylinder pairs as a function of position as well as the average drag coefficient (CDv) of larger arrays. A strain gauge was used to measure CD whilst CDv was computed with a momentum balance which was validated by strain gauge measurements for a regularly spaced array. The velocity and turbulence intensity surrounding a pair of cylinders arranged one behind the other with respect to mean flow (in tandem) were also measured with an Acoustic Doppler Velocimeter. The isolated cylinder CD was found to be 2.11 in close agreement with other researchers. Under fixed flow conditions CD for a cylinder in a pair was found to be as low as -0.40 and as high as 3.46 depending on their relative positioning. For arrays, CDv was influenced more by the distribution of cylinders than the flow conditions over the range of conditions tested. Mean values of CDv for each array were found to be between 1.52 and 3.06. This new insight therefore suggests that CDv for vegetation in bulk may actually be much higher than the typical value of 1 which is often assumed to apply in practice. If little other information is available, a crude estimate of CDv = 2 would be reasonable for many practical applications. The validity of a 2D realizable k-epsilon turbulence model for predicting the flow around square cylinders was evaluated. The model was successful in predicting CD for an isolated cylinder. In this regard the model performed as well as Large Eddy Simulations by other authors with a significant increase in computational efficiency. However, the numerical model underestimates CD of downstream cylinders in tandem pairs and overestimates velocities in their wake. This suggests it may be necessary to expand the model to three-dimensions when attempting to simulate the flow around two or more bluff obstacles with sharp edges.

Un grand effort a été réalisé ces dernières années dans la compréhension du champ d'écoulement et du transfert de chaleur dans les canaux de refroidissement internes présents dans les pales de turbine. En effet, des systèmes de refroidissement avancés ont non seulement conduit à l'augmentation de l'efficacité de la turbine à gaz en augmentant la température d'entrée de la turbine au-dessus de la température de fusion du matériau, mais également en augmentant la durée de vie de la turbine. Pour permettre de tels progrès, des techniques expérimentales et numériques modernes ont été largement appliquées afin d'interpréter et d'optimiser l'aérodynamique et le transfert de chaleur dans les canaux de refroidissement internes. Cependant, les données disponibles sont limitées dans le cas des canaux de refroidissement internes dans les aubes de rotor de turbine. Les gradients de rotation et de température introduisent des forces de flottabilité de type Coriolis et centripète dans le référentiel rotatif, modifiant de manière significative l'aérothermodynamique par rapport aux passages stationnaires. Dans le cas des pales de rotor de turbine, la plupart des investigations sont soit basées sur des mesures ponctuelles, soit sont contraintes à des régimes de rotation faibles. L'objectif principal de ce travail est d'étudier le débit détaillé et le transfert de chaleur d'un canal de refroidissement interne à des conditions de fonctionnement dimensionnelles sans moteur représentatives. Ce travail introduit une section d'essai en laboratoire qui exploite des canaux à nervures sur un large éventail de nombres de Reynolds, de rotation et de flottabilité. Dans le présent travail, le nombre de Reynolds va de 15,000 à 55,000, le nombre de rotation maximum est égal à 0.77 et le nombre maximal de flottabilité est égal à 0.77. La nouvelle installation expérimentale consiste en une conception polyvalente qui permet l'interchangeabilité de la géométrie testée, de sorte que les canaux de différents rapports d'aspect et les géométries de nervure peut être facilement installé. La particle image velocimetry et la thermographie à cristaux liquides sont effectuées pour fournir des mesures précises de vitesse et de transfert de chaleur dans les mêmes conditions opératoires, ce qui conduit à un ensemble de données expérimentales unique. De plus, des simulations à grands virages sont réalisées pour donner une image de l'ensemble du champ d'écoulement et compléter les observations expérimentales. En outre, l'approche numérique vise à fournir une méthodologie robuste qui est capable de fournir des prédictions haute-fidélité de la performance des canaux de refroidissement internes. / A great effort has been carried out over the recent years in the understanding of the flow field and heat transfer in the internal cooling channels present in turbine blades. Indeed, advanced cooling schemes have not only lead to the increase of the gas turbine efficiency by increasing the Turbine Inlet Temperature above the material melting temperature, but also the increase of the turbine lifespan. To allow such progresses, modern experimental and numerical techniques have been widely applied in order to interpret and optimize the aerodynamics and heat transfer in internal cooling channels. However, the available data is limited in the case of internal cooling channels in turbine rotor blades. Rotation and temperature gradients introduce Coriolis and centripetal buoyancy forces in the rotating frame of reference, modifying significantly the aerothermodynamics from that of the stationary passages. In the case of turbine rotor blades, most of the investigations are either based on point-wise measurements or are constraint to low rotational regimes. The main objective of this work is to study the detailed flow and heat transfer of an internal cooling channel at representative engine dimensionless operating conditions. This work introduces a laboratory test section that operates ribbed channels over a wide range of Reynolds, Rotation and Buoyancy numbers. In the present work, the Reynolds number ranges from 15,000 to 55,000, the maximum Rotation number is equal to 0.77, and the maximum Buoyancy number is equal to 0.77. The new experimental facility consists in a versatile design that allows the interchangeability of the tested geometry, so that channels of different aspect ratios and rib geometries can be easily fitted. Particle Image Velocimetry and Liquid Crystal Thermography are performed to provide accurate velocity and heat transfer measurements under the same operating conditions, which lead to a unique experimental data set. Moreover, Large Eddy Simulations are carried out to give a picture of the entire flow field and complement the experimental observations. Additionally, the numerical approach intends to provide a robust methodology that is able to provide high fidelity predictions of the performance of internal cooling channels.

[pt] Os modelos mais populares para solucionar escoamentos turbulentos são baseados no esquema RANS (Reynolds Average Navier Stokes) que necessita de fechamento para relacionar o tensor de tensões de Reynolds com os tensores médios cinemáticos. A solução clássica é a aproximação por Bussinesq que assume uma relação linear entre a parte deviatórica do Tensor de Reynolds e o tensor das taxas de deformação. Trabalhos anteriores mostraram que uma relação não linear entre o tensor das taxas de deformação pode melhorar a predição do modelo. No presente trabalho, primeiramente é realizada uma avaliação entre modelos lineares presentes na literatura seguido de uma análise de três modelos de ordem elevada que expandem a base tensorial para incluir tensores ortogonais. Duas adimensionalizações, uma com a energia cinética turbulenta e taxa de dissipação e outra com energia cinética turbulenta e intensidade do tensor de deformação, haviam sido propostas. As previsões dos modelos são comparados com dados DNS para um canal e para uma gama variada de número de Reynolds. Todos os modelos são implementados na plataforma aberta OpenFoam. Predições razoáveis para a componente cisalhante de todos os modelos foram obtidas quando comparadas com os dados DNS. Entretanto, modelos não lineares provaram superioridade na predição das outras componentes. Também foi observado que o modelo não linearmente dependente do tensor taxa de deformação e o tensor não persistencia das deformações foi o que melhor representou os campos providos por DNS. / [en] The most popular models to solve turbulent flows are based on the Reynolds Average Navier Stokes approach (RANS), which needs closure equations to relate the Reynolds stress tensor to the mean kinematic tensors. The classical approach is the Boussinesq approximation that assumes a linear relation between the deviatoric part of the Reynolds stress tensor, and the rate of strain tensor. Previous works have shown, that the non-linear dependence on the rate of strain tensor can improve the model predictions. At the present work, first an evaluation of linear models available in the literature is performed, followed by the analysis of three higher order methods, that expands the tensorial basis to include other objective orthogonal tensors. Two different nondimensionalization, one with the turbulent kinetic energy and dissipation rate and the other one with turbulent kinetic energy and the intensity of the rate of strain, had also been proposed for the models. The performance of the new models is assessed by comparing their numerical predictions to available channel flow and for a broad range of Reynolds Numbers. All models are implemented in the open source platform OpenFOAM. Reasonable predictions of the Reynolds shear component of all models were obtained when compared with the DNS data. However, the non-linear models proved superior in the prediction of the other components. It was also observed that the model which depends nonlinearly with the rate of strain and linearly with the non-persistence of strain was the one that best represented the DNS data field.

In some species, including humans, red blood cells (RBCs) under low shear stress tend to clump together and form into regular stacks called rouleaux. These stacks are not static, and constantly move and break apart. This phenomenon is referred to as red blood cell aggregation and disaggregation. When modelled as a single liquid, blood behaves as a non-Newtonian fluid. Its viscosity varies, mainly due to the aggregation of RBCs. The aim of this research is to develop a mesoscale computational model for the simulation of RBCs in plasma. This model considers RBC interaction and aggregation to predict blood-flow characteristics such as viscosity, rouleaux size and velocity distribution. In this work, the population-balance modelling (PBM) approach is utilized to model the RBC aggregation process. The PBM approach is a known method that is used for modelling agglomeration and breakage in two-phase flow fluid mechanics to find aggregate size. The PBM model is coupled to the incompressible Navier-Stokes equations for the plasma. Both models are numerically solved simultaneously. The population-balance equation has been used previously in a more restricted form, the Smoluchowski equation, to model blood viscosity, but it has never been fully coupled with the Navier-Stokes equation directly for the numerical modelling of blood flow. This approach results in a comprehensive model which aims to predict RBC aggregate size and their velocities for different flow configurations, as well as their effects on the apparent macro-scale viscosity. The PBM approach does not treat the microscopic physics of aggregation directly but rather uses experimental correlations for aggregation and disaggregation rates to account for the effects of aggregation on the bulk. To find the aggregation rate, a series of experiments on RBC sedimentation due to gravity is designed. In these tests, aggregated RBCs (rouleaux) tend to settle faster than single RBCs and, due to low shear stresses, disaggregation is very low and can be neglected. A high-speed camera is used to acquire video-microscopic pictures of the process. The size of the aggregates and their velocities are extracted using image processing techniques. For image processing, a general Matlab program is developed which can analyze all the images and report the velocity and size distribution of rouleaux. An experimental correlation for disaggregation rate is found using results from a previous steady-state Couette flow experiment. Aggregation and disaggregation rates from these experiments are used to complete the PBM model. Pressure-driven channel flow experiments are then used for the final validation of the model. Comparisons of the apparent viscosity of whole blood in previous experiments show reasonable agreement with the developed model. This model fills a gap between micro-scale and macro-scale treatments and should be more accurate than traditional macro-scale models while being cheaper than direct treatment of RBCs at the micro-scale.

La modélisation des phénomènes physiologiques induits par un écoulement, tels que l'écoulement sanguin au travers d'une sténose ou l'écoulement d'air lors de la production de parole, repose souvent sur des théories quasi-unidimensionnelles ou bi-dimensionnelles. Cependant, il est établi que le développement des couches limites dépend de la section transversale. Le but de cette thèse est de modéliser, simuler et caractériser l'importance potentielle de la section transversale sur les écoulements laminaires, contrôlés en pression, en l'absence ou en présence d'une constriction. Des coordonnées de translation sont utilisées pour obtenir des solutions pour des écoulement visqueux au travers d'une section de forme arbitraire. Cette paramétrisation est appliquée à la résolution des équations physiques pour des formes à deux et à trois dimensions. Un modèle d'écoulement simplifié quasi-tridimensionnel, qui prend en compte les pertes dissipatives par convection, la viscosité et la forme de la section est présenté et appliqué à la description de l'écoulement le long d'une sténose. Des données expérimentales et issues de simulations numériques sont collectées afin de caractériser l'influence de la forme de la section transversale dans le cas d'une constriction. simulation numérique sont comparées. / Physical models of physiological flow-induced phenomena, such as blood flow through a stenosis or air flow during human speech production, often rely on a quasi-one-dimensional or two-dimensional flow model, so that details of the cross section shape are neglected. Nevertheless, boundary layer development is known to depend on the cross section shape. The aim of this thesis is to model, simulate and characterize the potential impact of the cross section shape for pressure-driven laminar channel flow without and with constriction. Stretched coordinates are introduced to obtain viscous flow solutions for channels with an arbitrary cross section. The proposed cross section shape parametrization is applied to solve physical equations for two-dimensional and three-dimensional shapes. A simplified quasi-three-dimensional flow model, which accounts for kinetic losses, viscosity and the cross section shape, is presented and applied to describe the flow through a stenosis. Finally, flow data are gathered experimentally and numerically in order to characterize the influence of the cross section shape in the case of a constricted channel. Modeled, experimental and numerical data are compared.

Flow induced self-oscillations cause acoustic pressure oscillations of large amplitude in pipe flows. If Reynolds number is treated as a parameter, these floinduced oscillations occur only at discrete and critical values of Reynolds number. However, for a small range of Reynolds numbers around such a critical value, such periodic oscillations may appear intermittently. If intermittency, which is a precursor to these self-oscillations, can be detected, prediction of an impending instability may be possible. In experiments done by Vineeth and Sujith (Int. J. Aeroacoustics, 2016) on flow in a duct orifice arrangement, where flow enters through the duct inlet, and leaves into the atmosphere through the orifice exit, “whistling” was observed at a Reynolds number of 4200 (based on the orifice thickness and flow speed within the orifice), where large amplitude pressure oscillations were observed. At slightly lower Reynolds numbers, bursts of relatively smaller amplitudes of pressure oscillations were observed to appear intermittently. For a similar configuration, Large Eddy Simulations (LES) have been carried out with explicit filtering as a sub­ grid scale model here. Both whistling and intermittency are observed in the simulations. As air flows from the duct into the orifice, it turns sharply around the corner at the duct­ orifice interface. Due to this sharp turn, flow separation occurs, and hence, a shear layer is formed at the mouth of the orifice. The mechanism of whistling is found to be this shear layer within the orifice flapping about and hitting the trailing edge of the orifice periodically, thus causing the shear layer to break and roll up into a vortex. At Reynolds numbers where intermittency is observed, the shear layer is found to very mildly come in contact with the edges of the orifice walls, thus disturbing it. In the simulations, time series data of pressure are recorded at various probe locations. In a given time series, if scale invariance behaviour exists, it can be quantified by measuring the Hurst exponent. The numerical value of the Hurst exponent is an index of “long range or short range dependence” in a time series. Hurst exponent is measured in the time series data obtained. It is found to drop to zero as the flow approaches the state of a self-sustained oscillation, since the growth rates of all the long term as well as short term trends in the time series vanish. A loss of multifractality in the time series is also observed as the flow approaches whistling. As a part of the this thesis, new, split high resolution schemes of high order are designed following the Hixon Turmel Proposal.

Les écoulements turbulents de gaz denses, qui sont d’un grand intérêt pour un large éventail d'applications, sont le siège de phénomènes physiques encore peu connus et difficiles à étudier par des approches expérimentale. Dans ce travail, nous étudions pour la première fois l’influence des effets de gaz denses sur la structure de la turbulence compressible à l’aide de simulations numériques. Le fluide considéré est le PP11, un fluorocarbure lourd, dont le comportement thermodynamique a été représenté à l’aide de différentes lois d’état, afin de quantifier la sensibilité des solutions aux choix de modélisation. Nous avons considéré d’abord la décroissance d’une turbulence homogène isotrope compressible. Les fluctuations de température sont négligeables, alors que celles de la vitesse du son sont importantes à cause de leur forte dépendance de la densité. Le comportement particulier de la vitesse du son modifie de manière significative la structure de la turbulence, conduisant à la formation de shocklets de détente. L’analyse de la contribution des différentes structures à la dissipation d’énergie et à la génération d’enstrophie montre que, pour un gaz dense, les régions de forte dilatation jouent un rôle similaire à celles de forte compression, contrairement aux gaz parfaits, dans lesquels le comportement est fortement dissymétrique. Ensuite, nous avons mené des simulations numériques pour une configuration de canal plan en régime supersonique, pour plusieurs valeurs des nombres de Mach et de Reynolds. Les résultats confirment la validité de l’hypothèse de Morkovin. L’introduction d’une loi d’échelle semi-locale prenant en compte le variations de densité et viscosité, permet de comparer les profils des grandeurs turbulentes (contraintes de Reynolds, anisotropie, budgets d’énergie) avec ces observés en gaz parfait. Les variables thermodynamiques, quant à elles, présentent une évolution très différente pour un gaz parfait et pour un gaz dense, la chaleur spécifique élevée de ce dernier conduisant à un découplage des effets dynamiques et thermiques et à un comportement proche de celui d’un fluide incompressible avec des propriétés variables. / Dense gas turbulent flows, of great interest for a wide range of engineering applications, exhibit physical phenomena that are still poorly understood and difficult to reproduce experimentally. In this work, we study for the first time the influence of dense gas effects on the structure of compressible turbulence by means of numerical simulations. The fluid considered is PP11, a heavy fluorocarbon, whose thermodynamic behavior is described by means of different equations of state to quantify the sensitivity of solutions to modelling choices. First, we considered the decay of compressible homogeneous isotropic turbulence. Temperature fluctuations are found to be negligible, whereas those of the speed of sound are large because of the strong dependence on density. The peculiar behavior of the speed of sound significantly modifies the structure of the turbulence, leading to the occurrence of expansion shocklets. The analysis of the contribution of the different structures to energy dissipation and enstrophy generation shows that, for a dense gas, high expansion regions play a role similar to high compression ones, unlike perfect gases, in which the observed behaviour is highly asymmetric. Then, we carried out numerical simulations of a supersonic turbulent channel flow for several values of Mach and Reynolds numbers. The results confirm the validity of the Morkovin' hypothesis. The introduction of a semi-local scaling, taking into account density and viscosity variations across the channel, allow to compare the wall-normal profiles of turbulent quantities (Reynolds stresses, anisotropy, energy budgets) with those observed in ideal gases. Nevertheless, the thermodynamic variables exhibit a different evolution between perfect and dense gases, since the high specific heats of the latter lead to a decoupling of dynamic and thermal effects, and to a behavior close to that of variable property incompressible fluids.

In this dissertation, we are concerned with the numerical simulation for flows of electrically conducting fluids exposed to an external magnetic field (also known as magnetohydrodynamics or in short MHD). The aim of the present dissertation is twofold. First, the in-house CFD hydrodynamic solver SFELES is extended to MHD problems. Second, MHD turbulence is studied in the simple configuration of a MHD pipe flow within an external transverse magnetic field. Chapter 2 of this dissertation aims at reminding the physical equations that govern incompressible MHD problems. Two equivalent formulations are put forward in the particular case of quasi-static MHD. Chapter 3 is devoted to the detailed development of the hybrid spectral - stabilized finite element methods for quasi-static MHD problems. The extension of SFELES is made for both Cartesian and axisymmetric systems of coordinates. The short chapter 4 follows to provide the performances of SFELES executed by several processes in a parallel environment. The addition of a parallel direct solver is studied in regards with the memory and time requirements. The extension of SFELES is then validated in chapter 5 with test cases of increasing complexity. For this purpose, laminar flows with an existing analytical-asymptotic solution are considered. The subject of chapter 6 is the MHD turbulent pipe flow within an external transverse and uniform magnetic field. The results are partially compared with the corresponding hydrodynamic flow and with a few data available in the literature. / Le thème de cette thèse de doctorat est la simulation numérique d'écoulements de fluides conducteurs d'électricité qui sont exposés à un champ magnétique extérieur (également connu sous le nom de magnétohydrodynamique ou encore MHD). L'objectif de ce travail est double. Premièrement, le code CFD maison SFELES est étendu aux problèmes MHD. Deuxièmement, la turbulence MHD est étudiée dans la configuration de l'écoulement en conduite cylindrique à l'intérieur d'un champ magnétique transverse. Le chapitre 2 de cette thèse a pour but de rappeler les équations qui gouvernent les problèmes de MHD incompressible. Deux formulations équivalente sont mises en évidence dans le cas particulier de la MHD quasi-statique. Le chapitre 3 est dévoué au développement détaillé des méthodes spectrale - éléments finis pour la MHD quasi-statique. L'extension de SFELES est réalisée dans les systèmes de coordonnées cartésiennes et axisymétriques. Le court chapitre 4 suit pour fournir les performances de SFELES exécuté sur plusieurs processeurs dans un environnement parallèle. L'ajout d'un solveur parallèle direct est étudié en ce qui concerne les demandes en temps et mémoire. L'extension de SFELES est alors validée dans le chapitre 5 avec des cas d'étude de complexité croissante. Dans ce but, des écoulements laminaires avec solution théorique-asymptotique sont envisagés. Le sujet du chapitre 6 est l'écoulement MHD turbulent en conduite cylindrique à l'intérieur d'un champ magnétique transverse et uniforme. Les résultats sont partiellement comparés avec l'écoulement hydrodynamique correspondant et avec des données disponibles dans la littérature. / Doctorat en Sciences de l'ingénieur / info:eu-repo/semantics/nonPublished