Em diversos escoamentos sobre superfícies há a presença de protuberâncias, como por exemplo rebites, parafusos e juntas. Estas protuberâncias podem influenciar a camada limite, acelerando a transição do escoamento do estado laminar para o estado turbulento. Em alguns casos isto pode ser indesejável, já que o escoamento turbulento implica necessariamente em uma força de atrito maior do que aquela referente ao escoamento laminar. Existem alguns aspectos neste tipo de escoamento que ainda não estão bem compreendidos. O objetivo deste trabalho é estudar a influência de uma rugosidade isolada no escoamento sobre uma superfície. Este estudo contribui para se entender o que ocorre em casos de maior complexidade. O estudo é de natureza computacional, em que se utiliza simulação numérica direta das equações de Navier-Stokes. A técnica de fronteiras imersas é utilizada para representar a rugosidade no escoamento sobre a superfície. O código numérico é verificado por meio do método de soluções manufaturadas. Comparações entre resultados experimentais, da teoria de estabilidade linear e numéricos também são utilizados para a validação do código. Resultados obtidos com diferentes alturas de rugosidade e variações no gradiente de pressão permitiram analisar a influência de elemento rugoso tridimensional em escoamentos de camada limite. / The presence of protuberances on surfaces, for example, rivets, screws and gaskets, can influence the boundary layer by accelerating the transition from laminar flow to turbulent flow. In some cases this may be undesirable, since the turbulent flow involves frictional forces greater than the ones at the laminar regime. There are some aspects of the flow in the boundary layer perturbed by a single roughness element that are not well understood. The aim of this work is to study the influence of an isolated roughness on the boundary layer. This study is a step towards to the understanding of what can happen in more complex cases. The nature of this study is computational, therefore a Direct Numerical Simulation code is used. The immersed boundary method is used to represent the roughness in the flow on the surface. The numerical code is verified via theMethod ofManufactured Solutions. Comparisons between experimental data, Linear Stability Theory and numerical results are also used for the validation of the code. Results obtained with different roughness heights and variations in the pressure gradient allowed the analysis of the influence of a three-dimensional roughness element in boundary layer flows.

La propreté inclusionnaire reste un enjeu majeur en élaboration des métaux par voie liquide. La flottation, principal procédé retenu en métallurgie secondaire pour éliminer les particules d’inclusions, consiste à injecter des bulles de gaz au sein du réacteur. Lors de leur ascension, les bulles vont capter les plus grosses inclusions et favoriser la collision et l’agrégation des particules. Dans le but de quantifier les phénomènes influents à l’échelle des inclusions sur la dynamique d’agrégation entre deux inclusions à proximité des bulles, un modèle numérique 3D a été développé. L’écoulement local est modélisé par un cisaillement plan permanent et résolu par une méthode de Boltzmann sur réseau. Le couplage entre les particules et le fluide a été assuré par une méthode de frontière immergée permettant de calculer la perturbation hydrodynamique engendrée par la présence des particules et de mettre à jour les interactions entre particules pour leur suivi lagrangien. Les simulations numériques réalisées ont mis en évidence que les effets hydrodynamiques ont une influence non négligeable sur le comportement des inclusions. Des sections efficaces de collision ont pu être extraites, à partir desquelles ont été calculés des noyaux d’agrégation, données macroscopiques rendant compte des effets à petite échelle. Une première application de ce travail a été menée avec le calcul des fréquences d’agrégation d’un train de bulle dans un réacteur canal à partir de résultats de simulations DNS. Les noyaux d’agrégation ont également été exploités en vue de déterminer, à partir de résultats RANS de l’hydrodynamique d’une poche d’acier, l’évolution de la concentration d’inclusions par un bilan de population global. / Inclusion cleanliness remains a major challenge faced in process metallurgy in liquid phase. Flotation, the main process used in secondary metallurgy to remove inclusions, consists in injecting gas bubbles into the reactor. Rising gas bubbles entrap the biggest inclusions at their surface or in their wake. Besides, they promote collision and aggregation among particles. A 3D numerical model has been developed in order to quantify the roles of the prevailing phenomena on aggregation dynamics between inclusions in the vicinity of bubbles. At inclusion (mesoscopic) scale, the turbulent flow is locally modeled by a steady plane shear flow which is solved using a lattice-Boltzmann method. The coupling between both liquid and solid phases is ensured using an immersed boundary method. This method resolves the hydrodynamic perturbation induced by particles, and hence their interactions that are, in turn, used to update their Lagrangian tracking. The conducted numerical simulations bring out the influence of hydrodynamic effects on inclusion behavior. Collision cross sections have been determined from which ensuing aggregation kernels have been calculated. Such cross sections could provide macroscopic models to represent local particle dynamics. A first application of these results is presented to calculate aggregation frequencies in bubble swarms in a channel flow reactor that was simulated using DNS. Additionally, evolution of inclusion populations in molten steel has been determined from RANS simulation of a liquid steel ladle by means of a global population balance implementing the aggregation kernels determined in the present work.

This thesis presents a numerical method for the phase-resolving simulation of rigid particles and deformable bubbles in viscous, magnetohydrodynamic flows. The presented approach features solid robustness and high numerical efficiency. The implementation is three-dimensional and fully parallel suiting the needs of modern high-performance computing. In addition to the steps towards magnetohydrodynamics, the thesis covers method development with respect to the immersed boundary method which can be summarized in simple words by From rigid spherical particles to deformable bubbles. The development comprises the extension of an existing immersed boundary method to non-spherical particles and very low particle-to-fluid density ratios. A detailed study is dedicated to the complex interaction of particle shape, wake and particle dynamics. Furthermore, the representation of deformable bubble shapes, i.e. the coupling of the bubble shape to the fluid loads, is accounted for. The topic of bubble interaction is surveyed including bubble collision and coalescence and a new coalescence model is introduced. The thesis contains applications of the method to simulations of the rise of a single bubble and a bubble chain in liquid metal with and without magnetic field highlighting the major effects of the field on the bubble dynamics and the flow field. The effect of bubble coalescence is quantified for two closely adjacent bubble chains. A framework for large-scale simulations with many bubbles is provided to study complex multiphase phenomena like bubble-turbulence interaction in an efficient manner.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / In this work, the combined multi-direct forcing and immersed boundary method (IBM) were presented to simulate uid-structure interaction problems. The multi-direct forcing is used aim at satisfying the no-slip condition in the immersed boundary. For the numerical simulations was used a multi-purpose computer code that is being developed in the MFlab - Fluid Mechanics Laboratory of Federal University of Uberl^andia. Tests are made to validate the numerical schemes and routines were implemented to simulate uid-structures interaction problems. Furthermore, computational tools are developed to construct and manage and optimize the use of a Beowulf cluster where all the parallel simulations presented in this work were done. The Method of Manufactured Solutions has been used for order-of-accuracy verication in the computational uid dynamics code. Two uid-structure interaction problems were studied using this methodology. The rst is a ow over a sphere for some Reynolds numbers. The results were compared to empirical results, obtaining satisfactory approximations. The second one is a immersed simple pendulum. For this problem the results are in agreement with physics. Indeed, these are preliminar results. New tests must be done to make progress in the methodology. Improvements are proposed in the IBM, in the uid-structure model, in the turbulence model, in the method used to discretize the uid domain. It is also proposed to apply the methodology to real problems as risers and valves. / O presente trabalho tem como principal objetivo a aplicação do método multifoçagem (MMF) para solução numérica tridimensional de problemas de interação uidoestrutura, buscando-se garantir a condição de não-escorregamento na região da fronteira imersa. Para as simulações numéricas foi utilizado um código computacional multipropósito em desenvolvimento no MFlab - Laboratório de Mecânica dos Fluidos da Universidade Federal de Uberlândia. Foram feitas modificações nesse código para que se pudesse validá-lo para solução de problemas com fronteira imersa e foi implementada uma rotina para solução de um problema de interação uido-estrutura total. Além disso, foi desenvolvido um pacote de ferramentas computacionais que possibilitou instalar e melhorar o desempenho de um cluster do tipo Beowulf utilizado para o desenvolvimento das simulações num eriças em paralelo do presente trabalho. Utilizando o Método das Soluções Manufaturadas foram obtidas soluções sintetizadas para as equações de Navier-Stokes, o que possibilitou obter a ordem de convergência numérica do código computacional para problemas contínuos e a validação deste código para problemas envolvendo corpos imersos ao combinar a o método das soluções manufaturadas com a metodologia de fronteira imersa. Na sequência foi solucionado o problema de escoamento ao redor de uma esfera parada, cujos resultados foram comparados com referencias empíricas, obtendo-se boa aproximação. Ainda para esse caso foi feita a avalição da norma L2 para as soluções num eriças obtidas nos pontos lagrangianos verificando a garantia da condição de não-escorregamento e feita uma análise da inuência dos número de ciclos utilizados no método multi-forçagem. Foi vericado que a solução numérica obtida depende do número de ciclos o que faz com que seja necessário se estabelecer um critério de convergência para este método. Um segundo problema de interação uido-estrutura total foi estudado. Consiste em um pêndulo simples imerso em um uido que parte de uma dada posição angular inicial e oscila em torno da sua posição de equilíbrio, até parar. Para esse caso foram feitas análises quantitativas. Os resultados são preliminares mas coerentes com a física do problema, indicando que a metodologia é adequada para solução deste tipo de problema. / Mestre em Engenharia Mecânica

Em diversos escoamentos sobre superfícies há a presença de protuberâncias, como por exemplo rebites, parafusos e juntas. Estas protuberâncias podem influenciar a camada limite, acelerando a transição do escoamento do estado laminar para o estado turbulento. Em alguns casos isto pode ser indesejável, já que o escoamento turbulento implica necessariamente em uma força de atrito maior do que aquela referente ao escoamento laminar. Existem alguns aspectos neste tipo de escoamento que ainda não estão bem compreendidos. O objetivo deste trabalho é estudar a influência de uma rugosidade isolada no escoamento sobre uma superfície. Este estudo contribui para se entender o que ocorre em casos de maior complexidade. O estudo é de natureza computacional, em que se utiliza simulação numérica direta das equações de Navier-Stokes. A técnica de fronteiras imersas é utilizada para representar a rugosidade no escoamento sobre a superfície. O código numérico é verificado por meio do método de soluções manufaturadas. Comparações entre resultados experimentais, da teoria de estabilidade linear e numéricos também são utilizados para a validação do código. Resultados obtidos com diferentes alturas de rugosidade e variações no gradiente de pressão permitiram analisar a influência de elemento rugoso tridimensional em escoamentos de camada limite. / The presence of protuberances on surfaces, for example, rivets, screws and gaskets, can influence the boundary layer by accelerating the transition from laminar flow to turbulent flow. In some cases this may be undesirable, since the turbulent flow involves frictional forces greater than the ones at the laminar regime. There are some aspects of the flow in the boundary layer perturbed by a single roughness element that are not well understood. The aim of this work is to study the influence of an isolated roughness on the boundary layer. This study is a step towards to the understanding of what can happen in more complex cases. The nature of this study is computational, therefore a Direct Numerical Simulation code is used. The immersed boundary method is used to represent the roughness in the flow on the surface. The numerical code is verified via theMethod ofManufactured Solutions. Comparisons between experimental data, Linear Stability Theory and numerical results are also used for the validation of the code. Results obtained with different roughness heights and variations in the pressure gradient allowed the analysis of the influence of a three-dimensional roughness element in boundary layer flows.

Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-01-11T14:54:52Z No. of bitstreams: 1 andresricardovaldez.pdf: 6592036 bytes, checksum: 23a86a3d84f13bffa421f219e7e4501d (MD5) / Rejected by Fabíola Rubim (fabiola.rubim@ufjf.edu.br), reason: on 2018-01-12T11:05:10Z (GMT) / Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-01-12T11:46:32Z No. of bitstreams: 1 andresricardovaldez.pdf: 6592036 bytes, checksum: 23a86a3d84f13bffa421f219e7e4501d (MD5) / Rejected by Adriana Oliveira (adriana.oliveira@ufjf.edu.br), reason: Favor corrigir: Membro da banca: Filho, José Karam on 2018-01-23T14:01:35Z (GMT) / Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-01-23T14:06:58Z No. of bitstreams: 1 andresricardovaldez.pdf: 6592036 bytes, checksum: 23a86a3d84f13bffa421f219e7e4501d (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2018-01-23T14:22:22Z (GMT) No. of bitstreams: 1 andresricardovaldez.pdf: 6592036 bytes, checksum: 23a86a3d84f13bffa421f219e7e4501d (MD5) / Made available in DSpace on 2018-01-23T14:22:22Z (GMT). No. of bitstreams: 1 andresricardovaldez.pdf: 6592036 bytes, checksum: 23a86a3d84f13bffa421f219e7e4501d (MD5) Previous issue date: 2017-09-18 / Neste trabalho são apresentados aspectos de modelagem computacional para o estudo de Interação Fluido-Estrutura (FSI). Numericamente, o Método de Lattice Boltzmann (LBM) é usado para resolver a mecânica dos fluidos, em particular as equações de Navier-Stokes incompressíveis. Neste contexto, são abordados problemas de escoamentos complexos, caracterizado pela presença de obstáculos. A imposição das restrições na interface fluido-sólido é feita utilizando princípios variacionais, empregando o Princípio de Balanço de Potências Virtuais (PVPB) para obter as equações de Euler-Lagrange. Esta metodologia permite determinar as dependências entre carregamentos cinematicamente compatíveis e o estado mecânico adotado. Neste sentido, as condições de interface fluido-sólido são abordadas pelo Método de Fronteira Imersa (IBM) visando técnicas computacionais de baixo custo. A metodologia IBM trata o equilíbrio das equações na interface fluido-sólido através da interpolação entre os nós Lagrangianos (sólidos) e os nós Eulerianos (fluidos). Neste contexto, uma modificação desta estratégia que fornece soluções mais precisas é estudada. Para mostrar as capacidades do acoplamento LBM-IBM são apresentados vários experimentos computacionais que demonstram grande fidelidade entre as soluções obtidas e as soluções disponíveis na literatura. / This work presents computational modeling aspects for studying Fluid-Structure Interaction (FSI). The Lattice Boltzmann Method (LBM) is employed to solve the fluid mechanics considering the incompressible Navier-Stokes equations. The flows studied are complex due to the presence of arbitrary shaped obstacles. The obstacles alters the bulk flow adding complexity to the analysis. In this work the Euler-Lagrange equations are obtained employing the Principle of Virtual Power Balance (PVPB). Consequently, the functional dependencies between the mechanical state and every kinematic compatible loadings are established employing variational arguments. This modeling technique allows to study the fluid-solid boundary constraint. In this context the fluid-solid interface is handled employing the Immersed Boundary Method (IBM). The IBM deals with the fluid-solid interface equilibrium equations performing an interpolation of forces between Lagrangian nodes (solid domain) and Eulerian Lattice grid (fluid domain). In this work a different version of this methodology is studied that allows to obtain more accurate solutions. To show the capabilities of the implemented LBM-IBM solver several experiments are done showing the agreement with the benchmarks results available in literature.

Une méthode numérique précise et efficace est proposée pour la simulation de corps déformables interagissant avec un écoulement incompressible. Les équations de Navier-Stokes, considérées dans leur formulation vorticité fonction de courant, sont discrétisées temporellement et spatialement à l'aide respectivement d'un schéma d'ordre 4 de Runge-Kutta et par des différences finies compactes. Grâce à l'utilisation d'un maillage uniforme, nous proposons un nouveau solveur direct au quatrième ordre pour l'équation de Poisson, permettant de garantir l'incompressibilité au zéro machine sur une grille optimale. L'introduction d'un corps déformable dans l'écoulement de fluide est réalisée au moyen d'une méthode de pénalisation de volume. La déformation du corps est imposée par l'utilisation d'un maillage lagrangien structuré mobile qui interagit avec le fluide environnant en raison des forces hydrodynamiques et du moment (calculés sur le maillage eulérien de référence). Une loi de contrôle efficace de la courbure d'un poisson anguilliforme nageant vers une cible prescrite est proposée. La méthode numérique développée prouve son efficacité et précision tant dans le cas de la nage du poisson mais aussi plus d'un grand nombre de problèmes d'interactions fluide-structure. / We present an efficient algorithm for simulation of deformable bodies interacting with two-dimensional incompressible flows. The temporal and spatial discretizations of the Navier--Stokes equations in vorticity stream-function formulation are based on classical fourth-order Runge--Kutta and compact finite differences, respectively. Using a uniform Cartesian grid we benefit from the advantage of a new fourth-order direct solver for the Poisson equation to ensure the incompressibility constraint down to machine zero over an optimal grid. For introducing a deformable body in fluid flow, the volume penalization method is used. A Lagrangian structured grid with prescribed motion covers the deformable body which is interacting with the surrounding fluid due to the hydrodynamic forces and the torque calculated on the Eulerian reference grid. An efficient law for controlling the curvature of an anguilliform fish, swimming toward a prescribed goal, is proposed which is based on the geometrically exact theory of nonlinear beams and quaternions. Validation of the developed method shows the efficiency and expected accuracy of the algorithm for fish-like swimming and also for a variety of fluid/solid interaction problems.

L’érosion éolienne est un phénomène complexe avec des interactions entre la couche limite atmosphérique, le transport des particules et la déformation des dunes. Dans cette thèse des simulations numériques de transport de particules solides sur des dunes fixes ou déformables sont effectuées. L’écoulement turbulent est calculé par des simulations des grandes échelles (LES) couplée avec une méthode de frontières immergées. Les particules solides sont tractées par une approche Lagrangienne. L’entraînement des particules, leur interaction avec la surface et leur dépôt sont pris en compte par des modèles physiques complets d’érosion. D’un point de vue numérique, une méthode de frontières immergées a été introduite pour simuler les écoulements turbulents sur des frontières mouvantes. Le nouveau solveur a été validé en effectuant des comparaison avec les résultats expérimentaux de Simoens et al. (2015) dans le cas d’une colline Gaussienne. D’un point de vue physique, des modèles complets ont été développés pour l’érosion éolienne en se basant sur les forces agissant sur les particules. Des modèles instantanés pour l’envol, le roulement et le glissement des particules sont développés pour initier le mouvement des particules. Leur rebond et le splash sont également pris en compte. Des équations Lagrangiennes sont utilisées pour simuler la trajectoire des particules solides dans l’air. Une équation de transport d’un lit de particules a également été développée pour les cas de glissement et de roulement des particules sur la surface. La déformation de la dune est effectuée en faisant le bilan des particules qui s’envolent et se déposent. Ces modèles ont été validés en comparant les résultats de simulation avec les résultats expérimentaux de Simoens et al. (2015) sur les profils de concentration autour d’une colline Gaussienne. Enfin, des simulations numériques d’une dune sinusoïdale déformable sont effectuées. La forme de la dune simulée est comparée avec les résultats expérimentaux de Ferreira and Fino (2012). Un bon agrément est obtenu a t = 2.0 min, par contre la hauteur de la dune est sous-estimée entre 4.0 min et 6.0 min. Les résultats numériques montrent que la zone de recirculation diminue progressivement quand la dune se déforme. L’érosion, due à l’envol et au splash, est important a l’avant de la dune tandis que les particules se déposent a l’arrière de la dune. Le modèle de splash a été modifié pour prendre en compte l’effet de la pente, ce qui a permis une meilleure estimation de la hauteur de la dune a t = 4.0 min. / Wind erosion is a complex dynamic process consisting in an atmospheric boundary layer, aeolian particle transport, sand dune deformation and their intricate interactions. This thesis undertakes this problems by conducting three-dimensional numerical simulations of solid particle transport over a fixed or deformable sand dune. Turbulent flow is calculated by a developed numerical solver (Large-eddy simulation (LES) coupled with immersed boundary method (IBM)). Solid particle trajectories are tracked by a Lagrangian approach. Particle entrainment, particle-surface interactions and particle deposition are taken into account by physical comprehensive wind erosion models. Firstly, a new numerical solver has been developed to simulate turbulent flows over moving boundaries by introducing the IBM into LES. Two canonical simulation cases of a turbulent boundary layer flow over a Gaussian dune and over a sinusoidal dune are performed to examine the accuracy of the developed solver. Recirculation region characteristics, mean streamwise velocity profiles, Reynolds stress profiles as well as the friction velocity over the dune are presented. In the Gaussian case, a good agreement between experimental data and simulated results demonstrates the numerical ability of the improved solver. In the sinusoidal case, the developed solver with wall modeling over the immersed boundary shows a better performance than the pure one, when a relatively coarse grid is used. Secondly, physical comprehensive modeling of wind erosion is described in detail, based on the forces acting an individual particle. An instantaneous entrainment model for both lifting and rolling-sliding modes is proposed to initialize particle incipient motions. Lagrangian governing equations of aeolian particle motion are presented and used to simulate the trajectories of solid particles. Particularly, Lagrangian governing equations of bed-load particle motion are originally deduced and applied to model the particle rolling-sliding movement on the bed surface. In addition, particle-surface interactions are taken into account by probabilistic rebound/splash models. Thirdly, numerical simulations of particle transport over a fixed Gaussian dune and over a deformable sinusoidal dune are carried out. In the fixed Gaussian case, an overall good agreement on the particle concentration profiles over the dune between the simulated results and the experimental data of Simoens et al. (2015) preliminarily validates the ability and accuracy of the developed numerical solver coupled with physical comprehensive wind erosion models. In the deformable sinusoidal case, the simulated dune shapes are compared with the experimental ones of Ferreira and Fino (2012). A good agreement between them is observed at t = 2.0 min and an obvious underestimate of the dune shape is shown at t = 4.0 min and t = 6.0 min. By analyzing the simulated results, it is shown that the recirculation zone behind the dune is gradually reduced as the dune deforms and that windward erosion and lee side deposition is observed. It is also shown after testing that the splash entrainment is important for the lee side erosion. Moreover, a preliminary attempt is presented to apply an improved splash model with accounting for the bed slope effect to the simulation of sand dune deformation. A better performance on the simulated dune shape is achieved at t = 4.0 min in comparison with the experimental one.

Les travaux menés, depuis plusieurs années, au CORIA ont abouti à la construction d’un outil numérique (ARCHER) permettant la simulation numérique directe d’écoulements diphasiques et notamment l’atomisation d’un jet liquide à haute vitesse. Ce type de simulation permet de capturer les phénomènes d’atomisation au voisinage de l’injecteur difficilement caractérisables par les outils expérimentaux actuels. Ces simulations requièrent des conditions d’injection délicates à évaluer a priori car elles dépendent des caractéristiques de l’écoulement au sein de l’injecteur. Or, certains jets présentent une grande sensibilité à ces conditions d’injection. Dès lors, il est nécessaire de simuler l’écoulement au sein de l’injecteur afin d’appréhender la nature de cette sensibilité. L’utilisation d’un maillage cartésien par le code ARCHER conjuguée à la volonté de simuler le système d’atomisation dans son ensemble ont orienté ces travaux vers l’utilisation d’une méthode de frontière immergée. Ces travaux ont ainsi permis de reproduire des écoulements au sein d’injecteurs de forme quelconque tout en conservant le maillage cartésien d’origine, précieux tant pour l’efficacité du solveur que pour sa précision. Dans un premier temps, l’implantation dans le code ARCHER d’une méthode de frontière immergée a été réalisée et testée sur des configurations de canal et de conduite et de l’écoulement autour d’un cylindre. L’application de cette méthode a porté sur la simulation de l’écoulement au sein d’un injecteur triple disque mono-trou et a notamment permis de caractériser l’origine de l’écoulement secondaire formé dans l’orifice de décharge. Afin d’évoluer vers la construction d’un outil numérique capable de simuler le système d’atomisation dans son ensemble, un couplage entre la méthode de frontière immergée et la méthode Ghost fluid a été nécessaire. La version bi-dimensionnelle développée a été testée sur la relaxation d’une goutte posée sur une paroi. Cette version a permis de simuler des écoulements au sein de canaux à différents rapports de longueur sur diamètre et l’écoulement au sein d’une buse convergente. La simulation simultanée de l’écoulement interne et externe a permis de lier les fluctuations de vitesses des écoulements internes à la création de surface engendrée sur les écoulements externes. / Since several years, the research conducted at the CORIA laboratory led to the development of a numerical tool (ARCHER) alllowing direct numerical simulations of two phase flows. In particular, the simulations of high speed liquid jet primary break-up have been strongly investigated. These simulations are able to capture primary break-up phenomena near the nozzle exit where experimental characterisations are difficult to conduct. These simulations need injection conditions tricky to gauge a priori, since they depend on the flow characteristics inside the nozzle. Moreover, some jets are highly sensitive to these injection conditions. Therefore, it becomes necessary to simulate the flow inside the nozzle to better understand this sensitive nature. The objective to simulate the whole atomization system guided the present work dedicated to the use of an immersed boundary method (IBM). Such an approach allows reproducing flows inside nozzles of arbitrary shape while keeping the original cartesian mesh valuable for numerical efficiency and accuracy. As a first step, the implementation of an IBM in ARCHER was carried out and tested on channels, pipes and uniform flows past a circular cylinder. An industrial application focused on the flow inside a triple disk compound injector. This work led to a refined description of the secondary flow origin in the discharge hole. In order to move towards the design of a numerical tool able to simulate the whole injection system, a coupling between IBM and the Ghost Fluid Method (GFM) has been found necessary. This allows accounting for two phase flows inside the nozzle where the dynamics of the triple line has to be considered. The bidimensional developments have been tested on drops released on walls. This version enabled to simulate flows inside channels with different ratios of length over diameter and the flow inside a convergent nozzle. The simultaneous computation of flows inside and outside nozzle has enabled to link the velocity fluctuations of internals flows to the surface setting-up gene-rated on external flows.

This thesis presents a numerical method for the simulation of fluid-structure interaction (FSI) problems on high-performance computers. The proposed method is specifically tailored to interactions between Newtonian fluids and a large number of slender viscoelastic structures, the latter being modeled as Cosserat rods. From a numerical point of view, such kind of FSI requires special techniques to reach numerical stability. When using a partitioned fluid-structure coupling approach this is usually achieved by an iterative procedure, which drastically increases the computational effort. In the present work, an alternative coupling approach is developed based on an immersed boundary method (IBM). It is unconditionally stable and exempt from any global iteration between the fluid part and the structure part. The proposed FSI solver is employed to simulate the flow over a dense layer of vegetation elements, usually designated as canopy flow. The abstracted canopy model used in the simulation consists of 800 strip-shaped blades, which is the largest canopy-resolving simulation of this type done so far. To gain a deeper understanding of the physics of aquatic canopy flows the simulation data obtained are analyzed, e.g., concerning the existence and shape of coherent structures.