Neste trabalho, desenvolve-se um método de fronteira imersa para o estudo de escoamento compressível modelado pelas equações de Euler bidimensionais. O método de discretização de diferenças finitas é empregado, usando o método de Steger-Warming de ordem dois para discretizar as variáveis espaciais e o esquema de Runge-Kutta de ordem quatro para discretizar as variáveis temporais. O método da fronteira imersa foi empregado para o estudo de aeroelasticidade computacional em uma seção típica de aerofólio bidimensional com dois movimentos prescritos: torsional e vertical, com o objetivo de se verifcar a eficiência do método e sua aplicabilidade para problemas em aeroelasticidade computacional. Neste estudo desenvolveu-se também um programa de computador para simular escoamentos compressíveis de fluido invíscido utilizando a metodologia proposta. A verificação do código gerado foi feita utilizando o método das soluções manufaturadas e o problema de reflexão de choque oblíquo. A validação foi realizada comparando-se os resultados obtidos para o escoamento ao redor de uma seção circular e de uma seção de aerofólio NACA 0012 com os resultados experimentais, para cada caso. / In this work, an immersed boundary method is developed to study compressible flow modeled by the two-dimensional Euler equations. The finite difference method is employed, using the second order Steger-Warming method to discretizate the space variables and the fourth order Runge-Kutta method to discretizate the time variables. The immersed boundary method was employed to study computational aeroelasticity on a typical two-dimensional airfoil section with two prescribed motion: pitching and plunging, in order to verify the efficiency of the numerical method and its applicability in computational aeroelasticity problems. In this work, a computer program was developed to simulate compressible flows for inviscid fluids using the methodology proposed. The verification of the computational code was performed using the method of manufactured solutions and the oblique shock wave reflection problem. The validation was performed comparing the obtained results for flows around a circular section and a NACA 0012 airfoil section with the experimental results, for each case.

No description available.

Avec la hausse des niveaux de dimensionnement sismique il est devenu nécessaire de limiter les chargements internes dans les structures, notamment en utilisant des dispositifs glissants. Ces dispositifs plafonnent les efforts internes en déclenchant un glissement de la structure. Il devient cependant nécessaire d'estimer l'amplitude des déplacements de corps rigide, notamment pour les structures stockées dans des réservoirs. Dans ce cas, il est nécessaire de prévenir les impacts entre la structure glissante et les bords du réservoir pour contrôler les risques de fuite. Parmi les structures glissantes immergées, on citera les ponts, les structures côtières en maçonnerie, les râteliers de stockage de combustible nucléaire, etc...Les équations de dynamique associées au comportement de ces structures sont non-linéaires et nécessitent l'utilisation de simulations numériques coûteuses en temps de calcul et ne permettant pas de faire des études de sensibilité rapides. On propose donc une méthode de résolution quasi-analytique de ces équations en traitant dans un premier temps, l'évaluation analytique des matrices de masses ajoutées du couplage fluide-structure, dans un second temps, une méthode de résolution quasi-analytique du glissement d'une structure quelconque immergée dans un fluide avec une actualisation de la géométrie de lames d'eau. Les résultats obtenus présentent une bonne adéquation avec des simulations numériques et offrent un temps de calcul quasiment instantané compatible avec une étude paramétrique ou stochastique de ces structures / As the seismic loadings are increasing in accordance to the recent regulations regarding Earthquake design, the use of sliding devices in structures is becoming more common. These devices limitate the internal forces by creating a rigid body sliding. It is then necessary to estimate the global displacement of the structure, especially concerning structures that are immersed in a reservoir. In this case, the displacement must be well estimated in order to prevent impacts between the sliding structure and the boundaries of the reservoir. We can find such structures in : bridges, costal structures in brick and masonry, or in the nuclear industry with the underwater fuel storage racks, ...The governing equations for the behaviour of these structures are non linear and must be solved using time-consuming computer simulations which are not fit for a stochastic study. Our method consists in, firstly, evaluating analytically the added masses of the fluid-structure interaction, secondly, a semi-analytical solving of the governing equations including the updating of the dimensions of the fluid layers surrounding the sliding structure. The results of this new method are in accordance with the numerical simulations and can be obtained in a short time (1 or 2 seconds) which offers the possibility to make a stochastic analysis of the non linear behaviour

No desenvolvimento de códigos paralelos, a biblioteca PETSc se destaca como uma ferramenta prática e útil. Com o uso desta ferramenta, este trabalho apresenta um estudo sobre resolvedores de sistemas lineares aplicados a escoamentos incompressíveis de fluidos em microescala, além de uma análise de seu comportamento em paralelo. Após um estudo dos diversos aspectos dos métodos de fronteira imersa, é apresentado um método de fronteira imersa paralelo de primeira ordem. Na sequência, é apresentada uma proposta de melhoria na precisão do método, baseada na minimização da distância entre a condição de contorno exata e aproximada, no sentido de mínimos quadrados. O desenvolvimento de uma ferramenta paralela eficiente é demonstrado na solução numérica de problemas envolvendo escoamentos incompressíveis de fluidos viscosos com fronteiras imersas / In the development of parallel codes, PETSc library has an important position as a practical and useful tool. With this tool, this work presents a study about linear system solvers applied to incompressible flow in microscale problems, furthermore an analysis of the parallel behavior of these methods is presented. After a study of several aspects of immersed boundary methods, and taking advantage of the flexibility of PETSc, a parallel first order immersed boundary method is presented. Thereafter, an improvement in the accuracy of the method is presented, based on the minimization of the distance between exact and approximated boundary conditions, in the least square sense. The development of a parallel and efficient tool is demonstrated in the numerical solution of incompressible viscous flow problems with immersed boundary

Ce travail est dédié à la simulation d’écoulements multidimensionnels de gaz raréfiés dans un domaine où l’interface avec le solide est mobile. Le comportement du gaz est modélisé par un modèle de type BGK de l’équation de Boltzmann et une méthode déterministe de vitesses discrètes est utilisée pour discrétiser l’espace des vitesses microscopiques.Dans ce document, nous proposons tout d’abord trois discrétisations spatiales du modèle qui permettent la prise en compte du mouvement des parois solides, grâce à un traitement spécifique des conditions aux limites. Ces approches sont implémentées et validées pour plusieurs cas unidimensionnels et à la suite de cette étude, la méthode maille coupée est choisie pour une extension à des écoulements de dimensions plus élevées.La suite du travail présente l’algorithme utilisé pour la simulation d’écoulements 2D et 3D. La précision et la robustesse de l’implémentation sont mises en avant grâce à la simulation de nombreux cas tests, dont les résultats sont comparés à ceux issus de la littérature. La méthode maille coupée a notamment été optimisée par une technique de raffinement de maillage adaptatif. La simulation instationnaire 3D de la rotation des pâles du radiomètre de Crookes illustre pleinement le potentiel de la méthode. / This work is devoted to the multidimentional simulation of rarefied gases in a domain with moving boundary. The governing equation is given by BGKtype model of Boltzmann equation and velocity space is discretized with a standard discrete velocity method.We first propose three space discretizations that take boundary motion into account by specific treatment of the boundary conditions. These approaches are implemented and validated for several 1D flows. Based on this study, the cut cell method is chosen to be extend to multidimentional flows.Then we detail the cut cell algorithm for 2D and 3D flow simulations. Robustness and accuracy of the implementation are investigated through the simulation of numerous test cases. Our results are rigorously compared to the ones coming from the literature and good agreement is shown. The cut cell method has been optimized with an adaptive refinement mesh technique. The 3D unstationary simulation of the Crookes radiometer rotating vanes is a perfect illustration of the method potential.

Despite the high cost of memory and CPU time required to resolve the boundary layer, a viscous unstructured grid solver has many advantages over a structured grid solver such as the convenience in automated grid generation and shock or vortex capturing by solution adaption. Since the geometry and flow phenomenon of a helicopter are very complex, unstructured grid-based methods are well-suited to model properly the rotor-fuselage interaction than the structured grid solver. In present study, an unstructured Cartesian grid solver is developed on the basis of the existing solver, NASCART-GT. Instead of cut-cell approach, immersed boundary approach is applied with ghost cell boundary condition, which increases the accuracy and minimizes unphysical fluctuations of the flow properties. The standard k-epsilon model by Launder and Spalding is employed for the turbulence modeling, and a new wall function approach is devised for the unstructured Cartesian grid solver. It is quite challenging and has never done before to apply wall function approach to immersed Cartesian grid. The difficulty lies in the inability to acquire smooth variation of y+ in the desired range due to the non-body-fitted cells near the solid wall. The wall function boundary condition developed in this work yields stable and reasonable solution within the accuracy of the turbulence model. The grid efficiency is also improved with respect to the conventional method. The turbulence modeling is validated and the efficiency of the developed boundary condition is tested in 2-D flow field around a flat plate, NACA0012 airfoil, axisymmetric hemispheroid, and rotorcraft applications. For rotor modeling, an actuator disk model is chosen, since it is efficient and is widely verified in the study of the rotor-fuselage interaction. This model considers the rotor as an infinitely thin disk, which carries pressure jump across the disk and allows flow to pass through it. The full three dimensional calculations of Euler and RANS equations are performed for the GT rotor model and ROBIN configuration to test implemented actuator disk model along with the developed turbulence modeling. Finally, the characteristics of the rotor-fuselage interaction are investigated by comparing the numerical solutions with the experiments.

Single ventricle heart defects, where systemic and pulmonary venous returns mix in the single functional ventricle, represent the most complex form of congenital heart defect, affecting 2 babies per 1000 live births. Surgical repairs, termed "Fontan Repairs," reroute the systemic venous return directly to the pulmonary arteries, thus preventing venous return mixing and restoring normal oxygenation saturation levels. Unfortunately, these repairs are only palliative and Fontan patients are subjected to a multitude of chronic complications. It has long been suspected that hemodynamics play a role in determining patient outcome. However, the number of anatomical and functional variables that come into play and the inability to conduct large scale clinical evaluations, due to too small a patient population, has hindered decisive progress and there is still not a good understanding of the optimal care strategies on a patient-by-patient basis. Over the past decades, image-guided computational fluid dynamics (CFD) has arisen as an attractive option to accurately model such complex biomedical phenomena, providing a high degree of freedom regarding the geometry and flow conditions to be simulated, and carrying the potential to be automated for large sample size studies. Despite these theoretical advantages, few CFD studies have been able to account for the complexity of patient-specific anatomies and in vivo pulsatile flows. In this thesis, we develop an unstructured Cartesian immersed-boundary flow solver allowing for high resolution, time-accurate simulations in arbitrarily complex geometries, at low computational costs. Combining the proposed and validated CFD solver with an interactive virtual-surgery environment, we present an image-based surgical planning framework that: a) allows for in depth analysis of the pre-operative in vivo hemodynamics; b) enables surgeons to determine the optimum surgical scenario prior to the operation. This framework is first applied to retrospectively investigate the in vivo pulsatile hemodynamics of different Fontan repair techniques, and quantitatively compare their efficiency. We then report the prospective surgical planning investigations conducted for six failing Fontan patients with an interrupted inferior vena cava and azygous continuation. In addition to a direct benefit to the patients under consideration, the knowledge derived from these surgical planning studies will also have a larger impact for the clinical management of Fontan patients as they shed light onto the impact of caval offset, vessel flaring and other design parameters upon the Fontan hemodynamics depending on the underlying patient anatomy. These results provide useful surgical guidelines for each anatomical template, which could benefit the global surgical community, including centers that do not have access to patient-specific surgical planning interfaces.

La méthode IIM (Immersed Interface Method) permet d'étendre certaines méthodes numériques à des problèmes présentant des discontinuités. Elle est utilisée ici pour étudier un fluide incompressible régi par les équations de Navier-Stokes, dans lequel est immergée une membrane exerçant une force singulière. Nous utilisons une méthode de projection dans une grille de différences finies de type MAC. Une dérivation très complète des conditions de saut dans le cas où la viscosité est continue est présentée en annexe. Deux exemples numériques sont présentés : l'un sans membrane, et l'un où la membrane est immobile. Le cas général d'une membrane mobile est aussi étudié en profondeur. / The Immersed Interface Method allows us to extend the scope of some numerical methods to discontinuous problems. Here we use it in the case of an incompressible fluid governed by the Navier-Stokes equations, in which a membrane is immersed, inducing a singular force. We use a projection method and staggered (MAC-type) finite difference approximations. A very complete derivation for the jump conditions is presented in the Appendix, for the case where the viscosity is continuous. Two numerical examples are shown : one without a membrane, and the other where the membrane is motionless. The general case of a moving membrane is also thoroughly studied.

Fluid-structure interaction (FSI), occurrent in many areas of engineering and in the natural world, has been the subject of much research using a wide range of modelling strategies. However, problems with high levels of structural deformation are difficult to resolve and this is particularly the case for biomedical flows. A Lagrangian flow model coupled with a robust model for nonlinear structural mechanics seems a natural candidate since large distortion of the computational geometry is expected. Smoothed particle Hydrodynamics (SPH) has been widely applied for nonlinear interface modelling and this approach is investigated here. Biomedical applications often involve thin flexible structures and a consistent approach for modelling the interaction of fluids with such structures is also required. The Lagrangian weakly compressible SPH method is investigated in its recent delta-SPH form utilising inter-particle density fluxes to improve stability. Particle shifting is also used to maintain particle distributions sufficiently close to uniform to enable stable computation. The use of artificial viscosity is avoided since it introduces unphysical dissipation. First, solid boundary conditions are studied using a channel flow test. Results show that when the particle distribution is allowed to evolve naturally instabilities are observed and deviations are noted from the expected order of accuracy. A parallel development in the SPH group at Manchester has considered SPH in Eulerian form (for different applications). The Eulerian form is applied to the channel flow test resulting in improved accuracy and stability due to the maintenance of a uniform particle distribution. A higher-order accurate boundary model is developed and applied for the Eulerian SPH tests and third-order convergence is achieved. The well documented case of flow past a thin plate is then considered. The immersed boundary method (IBM) is now a natural candidate for the solid boundary. Again, it quickly becomes apparent that the Lagrangian SPH form has limitations in terms of numerical noise arising from anisotropic particle distributions. This corrupts the predicted flow structures for moderate Reynolds numbers (O(102)). Eulerian weakly compressible SPH is applied to the problem with the IBM and is found to give accurate and convergent results without any numerical stability problems (given the time step limitation defined by the Courant condition). Modelling highly flexible structures using the discrete element model is investigated where granular structures are represented as bonded particles. A novel vector-based form (the V-Model) is identified as an attractive approach and developed further for application to solid structures. This is shown to give accurate results for quasi-static and dynamic structural deformation tests. The V-model is applied to the decay of structural vibration in a still fluid modelled using Eulerian SPH with no artificial stabilising techniques. Again, results are in good agreement with predictions of other numerical models. A more demanding case representative of pulsatile flow through a deep leg vein valve is also modelled using the same form of Eulerian SPH. The results are free of numerical noise and complex FSI features are captured such as vortex shedding and non-linear structural deflection. Reasonable agreement is achieved with direct in-vivo observations despite the simplified two-dimensional numerical geometry. A robust, accurate and convergent method has thus been developed, at present for laminar two-dimensional low Reynolds number flows but this may be generalised. In summary a novel robust and convergent FSI model has been established based on Eulerian SPH coupled to the V-Model for large structural deformation. While these developments are in two dimensions the method is readily extendible to three-dimensional, laminar and turbulent flows for a wide range of applications in engineering and the natural world.

L'objectif de cette thèse est d'améliorer la compréhension des phénomènes jouant un rôle dans la mesure effectuée dans un analyseur sanguin, en particulier le comptage et la mesure de volumétrie d'une population de globules rouges reposant sur l'effet Coulter. Des simulations numériques sont effectuées dans le but de prédire la dynamique des globules rouges dans les zones de mesure et pour reproduire la mesure électrique associée, servant au comptage et à la volumétrie des cellules. Ces simulations sont effectuées à l'intérieur de configurations industrielles d'analyseur sanguin, en utilisant un outil numérique développé à l'IMAG, le solveur YALES2BIO. En utilisant la méthode des frontières immergées avec suivi de front, un modèle de particule déformable est introduit, celui-ci prend en compte le contraste de viscosité ainsi que les effets mécaniques de la courbure et de l'élasticité sur la membrane. Le solveur est validé grâce à de nombreux cas tests parcourant différents régimes et effets physiques. L'écoulement fluide dans cette géométrie d'analyseur sanguin est caractérisée par un fort gradient de vitesse axial dans la direction de l'écoulement, impliquant la présence d'un écoulement extensionnel au niveau du micro-orifice, là où a lieu la mesure. La dynamique des globules rouges est étudiée par des simulations numériques pour différentes conditions initiales, telles que sa position ou son orientation. Il est observé que les globules rouges vont se réorienter selon l'axe principal de l'analyseur sanguin dans tous les cas. Pour comprendre le phénomène, des modèles analytiques sont adaptés au cas des écoulements extensionnels et reproduisent correctement les tendances de réorientation.Cette thèse présente également la reproduction de la mesure électrique utilisée pour le comptage et la mesure de la distribution des volumes de globules rouges. De nombreuses simulations de la dynamique des globules rouges sont effectuées et utilisées pour générer l'impulsion électrique correspondant au passage du globule rouge dans le micro-orifice. Les amplitudes d'impulsions électriques résultantes permettent la caractérisation de la réponse électrique en fonction des paramètres initiaux de la simulation par une approche statistique. Un algorithme de Monte-Carlo est utilisé pour la quantification des erreurs de mesure liées à l'orientation et la position des globules rouges dans le micro-orifice. Ceci permet la génération d'une distribution de volume mesurée pour une population de globules rouges bien définie et la caractérisation des erreurs de mesure associées. / The aim of this thesis is to improve the understanding of the phenomena involved in the measurement performed in a blood analyzer, namely the counting and sizing of red blood cells based on the Coulter effect. Numerical simulations are performed to predict the dynamics of red blood cells in the measurement regions, and to reproduce the associated electrical measurement used to count and size the cells. These numerical simulations are performed in industrial configurations using a numerical tool developed at IMAG, the YALES2BIO solver. Using the Front-Tracking Immersed Boundary Method, a deformable particle model for the red blood cell is introduced which takes the viscosity contrast as well as the mechanical effects of the curvature and elasticity on the membrane into account. The solver is validated against several test cases spreading over a large range of regimes and physical effects.The velocity field in the blood analyzer geometry is found to consist of an intense axial velocity gradient in the direction of the flow, resulting in a extensional flow at the micro-orifice, where the measurement is performed. The dynamics of the red blood cells is studied with numerical simulations with different initial conditions, such as its position or orientation. They are found to reorient along the main axis of the blood analyzer in all cases. In order to understand the phenomenon, analytical models are adapted to the case of extensional flows and are found to reproduce the observed trends.This thesis also presents the reproduction of the electrical measurement used to count red blood cells and measure their volume distribution. Numerous dynamics simulations are performed and used to generate the electrical pulse corresponding to the passage of a red blood cell inside the micro-orifice. The resulting electrical pulse amplitudes are used to characterize the electrical response depending on the initial parameters of the simulation by means of a statistical approach. A Monte-Carlo algorithm helps quantifying the errors on the measurement of cell depending on its orientation and position inside the micro-orifice. This allows the generation of a measured volume distribution of a well defined red blood cell population and the characterization of the associated measurement errors.