Robust preconditioners on block-triangular and block-factorized form for three types of linear systems of two-by-two block form are studied in this thesis. The first type of linear systems, which are dense, arise from a boundary element type of discretization of crack propagation problems. Numerical experiment show that simple algebraic preconditioning strategies results in iterative schemes that are highly competitive with a direct solution method. The second type of algebraic systems, which are sparse, indefinite and nonsymmetric, arise from a finite element (FE) discretization of the partial differential equations (PDE) that describe (visco)elastic glacial isostatic adjustment (GIA). The Schur complement approximation in the block preconditioners is constructed by assembly of local, exactly computed Schur matrices. The quality of the approximation is verified in numerical experiments. When the block preconditioners for the indefinite problem are combined with an inner iterative scheme preconditioned by a (nearly) optimal multilevel preconditioner, the resulting preconditioner is (nearly) optimal and robust with respect to problem size, material parameters, number of space dimensions, and coefficient jumps. Two approaches to mathematically formulate the PDEs for GIA are compared. In the first approach the equations are formulated in their full complexity, whereas in the second their formulation is confined to the features and restrictions of the employed FE package. Different solution methods for the algebraic problem are used in the two approaches. Analysis and numerical experiments reveal that the first strategy is more accurate and efficient than the latter. The block structure in the third type of algebraic systems is due to a fine-coarse splitting of the unknowns. The inverse of the pivot block is approximated by a sparse matrix which is assembled from local, exactly inverted matrices. Numerical experiments and analysis of the approximation show that it is robust with respect to problem size and coefficient jumps.

The tips of unshrouded, high-pressure turbine blades are prone to significantly high heat loads. The gap between the tip and over-tip casing is the root cause of undesirable over-tip leakage flow that is directly responsible for high thermal material degradation and is a major source of aerodynamic loss within a turbine. Both must be minimised for the safe working and improved performance of future gas-turbines. A joint experimental and numerical study is presented to understand and characterise the heat transfer and aerodynamics of unshrouded blade tips. The investigation is undertaken with the use of a squealer or cavity tip design, known for offering the best overall compromise between the tip aerodynamics, heat transfer and mechanical stress. Since there is a lack of understanding of these tips at engine-realistic conditions, the present study comprises of a detailed analysis using a high-speed linear cascade and computational simulations. The aero-thermal performance is studied to provide a better insight into the behaviour of squealer tips, the effects of casing movement and tip cooling. The linear cascade environment has proved beneficial for its offering of spatially-resolved data maps and its ability to validate computational results. Due to the unknown tip gap height within an entire engine cycle, the effects of gap height are assessed. The squealer's aero-thermal performance has been shown to be linked with the gap height, and qualitative different trends in heat transfer are established between low-speed and high-speed tip flow regimes. To the author's knowledge, the present work is the first of its kind, providing comprehensive aero-thermal experimental research and a dataset for a squealer tip at engine-representative transonic conditions. It is also unique in terms of conducting direct and systematic validations of a major industrial computational fluid dynamics method for aero-thermal performance prediction of squealer tips at enginerepresentative transonic conditions. Finally, after recognising the highest heat loads are found on the squealer rims, a novel shaped squealer tip has been investigated to help improve the thermal performance of the squealer with a goal to improve its durability. It has been discovered that a seven percent reduction in tip temperature can be achieved through incorporating a shaped squealer and maximising the internal cooling performance.

Title: Numerical Solution of the Three-dimensional Compressible Flow Author: Martin Kyncl Department: Department of Numerical Mathematics Supervisor: Doc. RNDr. Jiří Felcman, CSc. Abstract: This thesis deals with a fluid flow in 3D in general. The system of the equations, describing the compressible gas flow, is solved numerically, with the aid of the finite volume method. The main purpose is to describe particular boundary conditions, based on the analysis of the incomplete Riemann problem. The analysis of the original initial-value problem shows, that the right hand-side initial condition, forming the Riemann problem, can be partially replaced by the suitable complementary condition. Several modifications of the Riemann problem are introduced and analyzed, as an original result of this work. Algorithms to solve such problems were implemented and used in code for the solution of the compressible gas flow. Numerical experiments documenting the suggested methods are performed. Keywords: compressible fluid flow, the Navier-Stokes equations, the Euler equations, boundary conditions, finite volume method, the Riemann problem, numerical flux, tur- bulent flow

Made available in DSpace on 2014-06-11T19:23:39Z (GMT). No. of bitstreams: 0 Previous issue date: 2007-12-07Bitstream added on 2014-06-13T20:11:13Z : No. of bitstreams: 1 lourenco_mas_me_ilha.pdf: 1600140 bytes, checksum: b00979a5a599fe5b08838113e8ca6489 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Este trabalho apresenta resultados da simulação numérica de escoamentos hipersônicos de fluidos, por meio de pySolver - um aplicativo computacional desenvolvido pelo autor. No aplicativo, as Equações de Euler foram discretizadas pelo método de elementos finitos de Galerkin (GFEM- Galerkin Finite Element Method) juntamente com a técnica CBS (Characteristic Based Split). O aplicativo pySolver, que foi construído baseado nas ferramentas de códigos fontes abertos Python, Blender e Visit, além da linguagem C, possui interface gráfica para o usuário, é multiplataforma e com orientação a objetos, além de contar com um framework especialmente projetado para a realização de todo o pré processamento, visando o modelamento geométrico bi ou tridimensional de problemas. O autor implementou um método para o refinamento de malha, modificando os programas abertos Triangle e TetGen, de tal forma a permitir o refinamento dinâmico e local de malhas até que determinadas tolerâncias sejam alcançadas nos resultados. Isto contribuiu para uma considerável robustez do aplicativo. Para verificação do aplicativo, foram simulados alguns casos-teste de escoamentos supersônicos e hipersônicos ao redor de corpo de diferentes configurações geométricas, principalmente aqueles encontrados na indústria aeronáutica e aeroespacial. Os dados obtidos são comparados com alguns resultados experimentais disponíveis na literatura, quando possível, e também com outros resultados numéricos obtidos da literatura. / This work presents some results for the numerical simulation of hypersonic fluid flows, utilizing pySolver – a software developed by the author. In this application, the Euler equations have been discretized by means of the Galerkin Finite Element Method (GFEM) using the CBS (Characteristic Based Split) scheme. pySolver, a multiplatform object-oriented software, built around the set of open source tools Python, Blender and Visit, besides C language, exhibits a proper graphical user interface and a framework specially developed to deal with data pre-processing and capable of geometrical modeling of either two or three-dimensional problems. The author has also implemented a scheme for the mesh refinement, by adapting the open-source softwares Triangle and TetGen, obtaining local and dynamic mesh refinement until reaching a determined tolerance in the results. That refinement scheme has contributed to considerable application robustness. In order to compare the software, some test cases composed of supersonic and hypersonic flows over di erent geometrical configuration bodies, mostly encountered in the aerospace and aeronautic industry data, have been simulated. The results compared very well with experimental data from the literature and, when possible, with other numerical results also obtained in the literature.

Resumo: Este trabalho apresenta resultados da simulação numérica de escoamentos hipersônicos de fluidos, por meio de pySolver - um aplicativo computacional desenvolvido pelo autor. No aplicativo, as Equações de Euler foram discretizadas pelo método de elementos finitos de Galerkin (GFEM- Galerkin Finite Element Method) juntamente com a técnica CBS (Characteristic Based Split). O aplicativo pySolver, que foi construído baseado nas ferramentas de códigos fontes abertos Python, Blender e Visit, além da linguagem C, possui interface gráfica para o usuário, é multiplataforma e com orientação a objetos, além de contar com um framework especialmente projetado para a realização de todo o pré processamento, visando o modelamento geométrico bi ou tridimensional de problemas. O autor implementou um método para o refinamento de malha, modificando os programas abertos Triangle e TetGen, de tal forma a permitir o refinamento dinâmico e local de malhas até que determinadas tolerâncias sejam alcançadas nos resultados. Isto contribuiu para uma considerável robustez do aplicativo. Para verificação do aplicativo, foram simulados alguns casos-teste de escoamentos supersônicos e hipersônicos ao redor de corpo de diferentes configurações geométricas, principalmente aqueles encontrados na indústria aeronáutica e aeroespacial. Os dados obtidos são comparados com alguns resultados experimentais disponíveis na literatura, quando possível, e também com outros resultados numéricos obtidos da literatura. / Abstract: This work presents some results for the numerical simulation of hypersonic fluid flows, utilizing pySolver - a software developed by the author. In this application, the Euler equations have been discretized by means of the Galerkin Finite Element Method (GFEM) using the CBS (Characteristic Based Split) scheme. pySolver, a multiplatform object-oriented software, built around the set of open source tools Python, Blender and Visit, besides C language, exhibits a proper graphical user interface and a framework specially developed to deal with data pre-processing and capable of geometrical modeling of either two or three-dimensional problems. The author has also implemented a scheme for the mesh refinement, by adapting the open-source softwares Triangle and TetGen, obtaining local and dynamic mesh refinement until reaching a determined tolerance in the results. That refinement scheme has contributed to considerable application robustness. In order to compare the software, some test cases composed of supersonic and hypersonic flows over di erent geometrical configuration bodies, mostly encountered in the aerospace and aeronautic industry data, have been simulated. The results compared very well with experimental data from the literature and, when possible, with other numerical results also obtained in the literature. / Orientador: João Batista Campos Silva / Coorientador: Emanuel Rocha Woiski / Banca: João Batista Aparecido / Banca: Paulo Gilberto de Paula Toro / Mestre

This thesis is concerned with the study of parallel mixing of two dissimilar gases under compressible conditions in the confined environment. A number of numerical studies are reported in the literature for the compressible mixing of two streams of gases where (1) both the streams are of similar gases at the same temperatures, (2) both the streams are at different temperatures with similar gases, and (3) dissimilar gases are with nearly equal temperatures. The combination of dissimilar gases at large temperature difference, mixing under compressible conditions, as in the case of scramjet propulsion, has not been adequately addressed numerically. Also many of the earlier studies have used two dimensional numerical simulation and showed good match with the experimental results on mixing layers that are inherently three dimensional in nature. In the present study, both two-dimensional (2-d) and three dimensional (3-d) studies are reported and in particular the effect of side wall on the three dimensionality of the flow field is analyzed, and the reasons of the good match of two dimensional simulations with experimental results have been discussed. Both two dimensional and three dimensional model free simulations have been conducted for a flow configuration on which experimental results are available. In this flow configuration, the mixing duct has a rectangular cross section with height to width ratio of 0.5. In the upper part of the duct hydrogen gas at a temperature of 103 K is injected through a single manifold of two Ludweig tubes and in the lower part of the duct nitrogen gas at a temperature of 2436 K is supplied through an expansion tube, both the gases are at Mach numbers of 3.1 and 4.0 respectively. Measurements in the experiment are limited to wall pressures and heat flux. The choice of this experimental condition gives an opportunity to study the effect of large temperature difference on the mixing of two dissimilar gases with large molecular weights under compressible conditions. Both two dimensional and three dimensional model free simulations are carried out using higher order numerical scheme (4th order spatial and 2nd order temporal) to understand the structure and evolution of supersonic confined mixing layer of similar and dissimilar gases. Two dimensional simulations are carried out by both SPARK (finite difference method) and OpenFOAM (finite volume method based open source software that was specially picked out and put together), while 3D model free simulations are carried out by OpenFOAM. A fine grid structure with higher grid resolution near the walls and shear layer is chosen. The effect of forcing of fluctuations on the inlet velocity shows no appreciable change in the fully developed turbulent region of the flow. The flow variables are averaged after the attainment of statistical steady state established through monitoring the concentration of inert species introduced in the initial guess. The effect of side wall on the flow structure on the mixing layer is studied by comparing the simulation results with and without side wall. Two dimensional simulations show a good match for the growth rate of shear layer and experimental wall pressures. Three dimensional simulations without side wall shows 14% higher growth rate of shear layer than that of two dimensional simulations. The wall pressures predicted by these three dimensional simulations are also lower than that predicted using two dimensional simulations (6%) and experimental (9%) results in the downstream direction of the mixing duct. Three dimensionality of the flow is thought of as a cause for these differences. Simulations with the presence of side wall show that there is no remarkable difference of three dimensionality of the flow in terms of the variables and turbulence statistics compared to the case without side walls. However, the growth rate of shear layer and wall surface pressures matches well with that predicted using two dimensional simulations. It has been argued that this good match in shear layer growth rate occurs due to formation of oblique disturbances in presence of side walls that are considered responsible for the decrease in growth rate in 3-d mixing layers. The wall pressure match is argued to be good because of hindrance from side wall in the distribution of momentum in third direction results in higher wall pressure. The effect of dissimilar gases at large temperature difference on the growth rate reduction in compressible conditions is studied. Taking experimental conditions as baseline case, simulations are carried out for a range of convective Mach numbers. Simulations are also carried out for the same range of convective Mach numbers considering the mixing of similar gases at the same temperature. The normalized growth rates with incompressible counterpart for both the cases show that the dissimilar gas combination with large temperature difference shows higher growth rate. This result confirms earlier stability analysis that predicts increased growth rate for such cases. The growth rate reduction of a compressible mixing layer is argued to occur due to reduced pressure strain term in the Reynolds stress equation. This reduction also requires the pressure and density fluctuation correlation to be very near to unity. This holds good for a mixing layer formed between two similar gases at same temperature. For dissimilar gases at different temperatures this assumption does not hold well, and pressure-density correlation coefficient shows departure from unity. Further analysis of temperature density correlation factor, and temperature fluctuations shows that the changes in density occur predominantly due to temperature effects, than due to pressure effects. The mechanism of density variations is found to be different for similar and dissimilar gases, while for similar gases the density variations are due to pressure variations. For dissimilar gases density variation is also affected by temperature variations in addition to pressure variations. It has been observed that the traditional k-ε turbulence model within the RANS (Reynolds Averaged Navier Stokes) framework fails to capture the growth rate reduction for compressible shear layers. The performance of k-ε turbulence model is tested for the mixing of dissimilar gases at large temperature difference. For the experimental test case the shear layer growth rate and wall pressures show good match with other model free simulations. Simulations are further carried out for a range of convective Mach numbers keeping the mixing gases and their temperatures same. It has been observed that a drop in the growth rate is well predicted by RANS simulations. Further, the compressibility option has been removed and it has been observed that for the density and temperature difference, even for incompressible case, the drop in growth rate exists. This behaviour shows that the decrease in growth rate is mainly due to the interaction of temperature and species mass fraction on density. Also it can be inferred that RANS with k-ε turbulence model is able to capture the compressible shear layer growth rate for dissimilar gases at high temperature difference. The mixing of heat and species is governed by the values of turbulent Prandtl and Schmidt numbers respectively. These numbers have been observed to vary for different flow conditions, while affecting the flow field considerable in the form of temperature and species distribution. Model free simulations are carried out on an incompressible convective Mach number mixing layer, and the results are compared with that of a compressible mixing layer to study the effect of compressibility on the values of turbulent Prandtl / Schmidt numbers. It has been observed that both turbulent Prandtl and Schmidt numbers show an almost constant value in the mixing layer region for incompressible case. While, for a compressible case, both turbulent Prandtl and Schmidt numbers show a continuous variation within the mixing layer. However, the turbulent Lewis number is observed to be near unity for both incompressible and compressible cases. The thesis is composed of 8 chapters. An introduction of the subject with critical and relevant literature survey is presented in chapter 1. Chapter 2 describes the mathematical formulation and assumptions along with solution methodology needed for the simulations. Chapter 3 deals with the two and three dimensional model free simulations of the non reacting mixing layer. The effect of the presence of side wall is studied in chapter 4. Chapter 5 deals with the effect of compressibility on the mixing of two dissimilar gases at largely different temperatures. The performance of k-ε turbulence model is checked for dissimilar gases in Chapter 6. Chapter 7 is concerned with the effect of compressibility on turbulent Prandtl and Schmidt numbers. Finally concluding remarks are presented in chapter 8. The main aim of this thesis is the exploration of parallel mixing of dissimilar gases under compressible conditions for both two and three dimensional cases. The outcome of the thesis is (a) a finding that the presence of sidewall in a mixing duct does not make flow field two dimensional, instead it causes the formation of oblique disturbances and the shear layer growth rate is reduced, (b) that it has been shown that the growth rates of dissimilar gases are affected far more by large temperature difference than by compressibility as in case of similar gases, (c) that the growth rates of compressible shear layers formed between dissimilar gases are better predicted using k-εturbulence model and (d) that for compressible mixing conditions the turbulent Prandtl and Schmidt numbers vary continuously in the mixing layer region necessitating the use of some kind of model instead of assuming constant values.

In this work, a DNS code was developed to investigate problems on transition in subsonic compressible boundary layer on a flat plate. Code validation tests were performed for linear and nonlinear stages of transition, on incompressible and compressible regimes. The focus of the present work is to investigate natural transition in subsonic boundary layers modeled by wave packets; and perform a preliminary study of transition induced by white noise. Three main problems were considered, namely, a DNS simulation and analysis of the ex- periment (MEDEIROS; GASTER, 1999b) of wave packet evolution on incompressible boundary layer, the influence of compressibility on wave packet evolution at subsonic Mach numbers and finally, a preliminary study of the evolution of a white noise perturbation in the boundary layer at Mach 0.2 and Mach 0.9. Comparisons between numerical and experimental results show remarkably good agreement in the linear and nonlinear stages, in both, spatial and Fourier spaces. A numerical simulation of this experiment and the analysis carried out is not available in the literature for wave packets in the incompressible boundary layer. The nonlinear modal analysis performed established the existence of tuned fundamental and subharmonic resonance of H-type and K-type in the packet. Influence of compressibility in the wave packet evolution was here investigated in boundary layers at Mach 0.7 and Mach 0.9. There are no works reported in the literature on wave packets in compressible subsonic boundary layer. In the linear regime, the oblique modes were the most unstable for Mach > 0.7, as expected by the results of the literature. In the nonlinear regime, strong streaks were observed, associated with low frequency modes that eventually decay downstream. An isolated wave packet at Mach 0.9 showed nonlinear amplification only in the subharmonic band, which may be associated to H-type or detuned resonance. However this packet has a relatively stable character. On the other hand, at Mach 0.9 spanwise interaction of wave packet pairs were more unstable than the isolated case, because stable modes for the isolated packet evolution becomes unstable in the wave packet interaction. This scenario evidenced the presence of oblique transition. Finally, the nonlinear evolution of the same white noise disturbance at Mach 0.2 and Mach 0.9 were observed to be completely different. In the incompressible boundary layer localized lambda vortex structures were observed, that could be associated to the local presence of H-type and/or K-type resonance. In the compressible regime, longitudinal vortex structures distributed across the entire domain seemed to be linked to oblique transition. In the white noise evolution, compressibility seems to have a stronger effect than in the wave packet evolution. In the conditions considered, the wave packet interaction appear to be a better representation of white noise compressible transition scenario. / No presente trabalho, um código DNS (Direct Numerical Simulation) foi desenvolvido para abordar problemas de transição para turbulência em camada limite subsônica compressível em uma placa plana. Foram realizados testes de validação de código , nos regimes linear e não linear do processo de transição, nos regimes incompressível e compressível. O foco do presente trabalho é estudar transição natural modelada por meio de pacotes de onda em camada limite compressível subsônica, e realizar uma análise preliminar da transição induzida por ruído branco. Três assuntos principais foram considerados: uma simulação DNS e uma análise comparativa com o experimento (MEDEIROS; GASTER, 1999b) sobre a evolução de um pacote de ondas em camada limite incompressível, a influência da compressibilidade na evolução de pacotes de ondas no regime subsônico, e por último, um estudo preliminar da transição induzida por ruído branco em Mach 0.2 e Mach 0.9. As comparações realizadas entre a solução numérica e os dados experimentais mostram uma boa concordância, nos regimes linear e não linear, tanto no espaço físico quanto no espaço de Fourier. A simulação numérica deste experimento e a análise realizada neste trabalho, não são encontradas na literatura para o regime incompressível. A análise modal não linear aplicada aos resultados, permitiu identificar a presença das ressonâncias tipo H e tipo K no pacote de ondas. A influência da compressibilidade na evolução dos pacotes de onda foi estudada em Mach 0.7 e Mach 0.9. Na literatura não há trabalhos sobre pacotes de ondas no regime sub- sônico. No regime linear da transição, os modos oblíquos resultam ser os mais instáveis para Mach > 0.7, como era de esperar, de acordo com os resultados da literatura. No regime não linear, foram observadas estrias de moderada amplitude, associadas com modos de baixa frequência que acabam decaindo. O pacote de ondas em Mach 0.9 apresentou amplificação não linear somente na banda subharmônica, que pode ser associada com transição tipo H ou ressonância dessintonizada. No entanto, o comportamento geral neste regime é estabilizante. Por sua vez, a interação entre pacotes de ondas em Mach 0.9 mostrou um comportamento desestabilizante, pois a interação acaba gerando amplificação não linear em modos que decaem no pacote isolado. Os modos amplificados sugerem a presença do mecanismo de transição oblíqua. Finalmente, a evolução da mesma perturbação constituída por ruído branco em Mach 0.2 e Mach 0.9, resultaram ser completamente diferentes. Na camada limite incompressível foram observados vórtices tipo lambda, que poderiam ser gerados pela presença localizada das ressonâncias tipo H e/ou tipo K. No regime compressível foram observados vórtices distribuidos em todo o domínio, o que sugere a presença da transição oblíqua. Na transição gerada por ruído branco a compressibilidade teve uma influência maior que no pacote de ondas. Nas condições estudadas, a interação entre pacotes de ondas parece ser uma melhor representação do ruído branco no regime compressível.

Neste trabalho foi realizado um estudo utilizando os modelos de turbulência Simulação das Grandes Escalas, Large Eddy Simulation (LES), e Simulação dos Vórtices Desprendidos, Detached Eddy Simulation (DES), para simular o escoamento compressível interno em um duto contendo válvulas controladoras na saída dos gases de combustão de um reator de craqueamento catalítico fluido, com o objetivo de comparar o desempenho numérico e computacional de ambas as técnicas. Para isso foi utilizado um programa comercial de dinâmica dos fluidos computacional, Computational Fluid Dynamics (CFD), que possui em seu código os dois modelos de turbulência. / In the present work a study was conducted using Large Eddy Simulation (LES) and Detached Eddy Simulation (DES) turbulence models in order to simulate the internal compressible flow in a duct containing the flue gas discharge control valves of a fluid catalytic cracking reactor so as to compare the numerical and computational behavior of both techniques. A commercial Computational Fluid Dynamics (CFD) software, which includes these turbulence models in its code, was used.

Cette thèse porte sur le développement de schémas semi-implicites à pas fractionnaires pour les équations de Navier-Stokes compressibles ; ces schémas entrent dans la classe des méthodes de correction de pression.La discrétisation spatiale choisie est de type "à mailles décalées :éléments finis mixtes non conformes (éléments finis de Crouzeix-Raviart ou Rannacher-Turek) ou schéma MAC classique.Une discrétisation en volumes finis décentrée amont du bilan de masse garantit la positivité de la masse volumique.La positivité de l'énergie interne est obtenue en discrétisant le bilan d'énergie interne continu, par une méthode de volumes finis décentrée amont, enfin, et en couplant ce bilan d'énergie interne discret à l'étape de correction de pression.On effectue une discrétisation particulière en volumes finis sur un maillage dual du terme de convection de vitesse dans le bilan de quantité de mouvement et une étape de renormalisation de la pression; ceci permet de garantir le contrôle au cours du temps de l'intégrale de l'énergie totale sur le domaine.L'ensemble de ces estimations a priori implique en outre, par un argument de degré topologique, l'existence d'une solution discrète. L'application de ce schéma aux équations d'Euler pose une difficulté supplémentaire.En effet, l'obtention de vitesses de choc correctes nécessite que le schéma soit consistant avec l'équation de bilan d'énergie totale, propriété que nous obtenons comme suit. Tout d'abord, nous établissons un bilan discret (local) d'énergie cinétique.Ce dernier comporte des termes sources, que nous compensons ensuite dans le bilan d'énergie interne. Les équations d'énergie cinétique et interne sont associées au maillage dual et primal respectivement, et ne peuvent donc être additionnées pour obtenir un bilan d'énergie totale ; cette dernière équation est toutefois retrouvée, sous sa forme continue, à convergence : si nous supposons qu'une suite de solutions discrètes converge lorsque le pas de temps et d'espace tendent vers 0,, nous montrons en effet, en 1D au moins, que la limite en satisfait une forme faible.Ces résultats théoriques sont confortés par des tests numériques.Des résultats similaires sont obtenus pour les équations de Navier-Stokes barotropes. / This thesis is concerned with the development of semi-implicit fractional step schemes, for the compressible Navier-Stokes equations; these schemes are part of the class of the pressure correction methods.The chosen spatial discretization is staggered: non conforming mixed finite elements (Crouzeix-Raviart or Rannacher-Turek) or the classic MAC scheme. An upwind finite volume discretization of the mass balanced guarantees the positivity of the density. The positivity of the internal energy is obtained by discretising the internal energy balance by an upwind finite volume scheme and by coupling the discrete internal energy balance with the pressure correction step.A special finite volume discretization on dual cells is performed for the convection term in the momentum balance equation, along with a renormalization of the pressure; this allows to guarantee the control in time of integral of the total energy over the domain.All these a priori estimates implies lead to the existence of a discrete solution by a topological degree argument.The application of this scheme the equations of Euler yields an additional difficulty.Indeed, obtaining correct shock speeds requires that the scheme be consistent with the total energy balance,, property which we obtain as follows.First of all, a local discrete kinetic energy balance is established; it contains source terms which are compensated by adding some source terms in the internal energy balance. The kinetic and internal energy equations are associated with the dual and primal meshes respectively, and thus cannot be added to obtain a balance total energy balance; its continuous counterpart is however recovered at the limit: if we suppose that a sequence of discrete solutions converges when the space and time steps tend to 0, we indeed show, in 1D at least, that the limit satisfies a weak form of the equation. These theoretical results are comforted by numerical tests.Similar results are obtained for the barotropic Navier--Stokes equations

Ce travail porte sur la modélisation de la formation et la dispersion d'un nuage de gouttes, par déconfinement d'un liquide: agression extérieure ou situation accidentelle. Le but est la construction d'un modèle apte à reproduire simultanément les conditions génératrices de la formation du nuage et l'évolution de ce nuage dans le temps (dispersion). La principale difficulté réside en la différence des modèles adaptés à la description d'écoulements caractérisant chaque étape du phénomène global : modèle d'écoulement multiphasique à phases compressibles (milieux continus) initialement, puis fragmentation et formation du nuage de gouttes dispersées dans une phase porteuse (modèle d'écoulements dilués). En l'absence de modèle analytique unique apte à décrire l'ensemble de ces processus, on propose une approche originale pour réaliser un couplage effectif entre ces deux modèles. La problématique de formation et de dispersion de liquide implique la prise en compte de plusieurs phénomènes physiques: fragmentation, transferts de chaleur et de masse ainsi que la traînée entre les phases. Ces différents phénomènes sont introduits dans le modèle global via des termes d'interactions présents dans les systèmes d'équations. La construction de ce modèle complet à permis la réalisation de calculs décrivant la formation et la dispersion d'un nuage de gouttes pouvant intervenir lors de situations accidentelles sur des sites industriels par exemple. / This work focuses on modeling the formation and the dispersion of a cloud of droplets, induced by ejection of a liquid, resulting from an external aggression or an accidental situation. The goal is to build a model able to reproduce simultaneously the conditions which generate the cloud formation and the cloud evolution in time (dispersion). The main difficulty lies in the differences between the already existing models adapted to the description of flows which are able to characterize each stage of the global phenomenon: initially a multiphase flow model with compressible phases (Continuum), then the atomization and the formation of a cloud of droplets dispersed in a carrier phase (dilute flow model). We propose a new approach to achieve an effective coupling between these two models. The problem of the formation and the dispersion of the liquid requires to take into account several physical phenomena: atomization, heat and mass transfers and drag between phases. These phenomena are included in the global model through interaction terms involved in the systems of equations. The construction of this model has permited the realization of calculations describing the formation and dispersion of a cloud of droplets which may occur during, for axample, in accidental situations at industrial sites.