This research is mainly focused on the development of the adaptive Cartesian grid methods for compressibl  e flow. At first, the ghost cell method and its applications for inviscid compressible flow on adaptive tree Cartesian grid are developed. The proposed method is successfully used to evaluate various inviscid compressible flows around complex bodies. The mass conservation of the method is also studied by numerical analysis. The extension to three-dimensional flow is presented. Then, an h-adaptive Runge–Kutta discontinuous Galerkin (RKDG) method is presented in detail for the development of high accuracy numerical method under Cartesian grid. This method combined with the ghost cell immersed boundary method is also validated by well documented test problems involving both steady and unsteady compressible flows over complex bodies in a wide range of Mach numbers. In addition, in order to suppress the failure of preserving positivity of density or pressure, which may cause blow-ups of the high order numerical algorithms, a positivity-preserving limiter technique coupled with h-adaptive RKDG method is developed. Such a method has been successfully implemented to study flows with the large Mach number, strong shock/obstacle interactions and shock diffraction. The extension of the method to viscous flow under the adaptive Cartesian grid with hybrid overlapping bodyfitted grid is developed. The method is validated by benchmark problems and has been successfully implemented to study airfoil with ice accretion. Finally, based on an open source code, the detached eddy simulation (DES) is developed for massive separation flow, and it is used to perform the research on aerodynamic performance analysis over the wing with ice accretion.

In this thesis, we study the Filippov moments solution for differential equations with discontinuous right-hand side. In particular, our aim is to define a suitable Filippov sliding vector field on a co-dimension $2$ manifold $\Sigma$, intersection of two co-dimension $1$ manifolds with linearly independent normals, and then to study the dynamics provided by this selection. More specifically, we devote Chapter 1 to motivate our interest in this subject, presenting several problems from control theory, non-smooth dynamics, vehicle motion and neural networks. We then introduce the co-dimension $1$ case and basic notations, from which we set up, in the most general context, our specific problem. In Chapter 2 we propose and compare several approaches in selecting a Filippov sliding vector field for the particular case of $\Sigma$ nodally attractive: amongst these proposals, in Chapter 3 we focus on what we called \emph{moments solution}, that is the main and novel mathematical object presented and studied in this thesis. There, we extend the validity of the moments solution to $\Sigma$ attractive under general sliding conditions, proving interesting results about the smoothness of the Filippov sliding vector field on $\Sigma$, tangential exit at first-order exit points, uniqueness at potential exit points among all other admissible solutions. In Chapter 4 we propose a completely new and different perspective from which one can look at the problem: we study minimum variation solutions for Filippov sliding vector fields in $\R^{3}$, taking advantage of the relatively easy form of the Euler-Lagrange equation provided by the analysis, and of the orbital equivalence that we have in the eventuality $\Sigma$ does not have any equilibrium points on it; we further remove this assumption and extend our results. In Chapter 5 examples and numerical implementations are given, with which we corroborate our theoretical results and show that selecting a Filippov sliding vector field on $\Sigma$ without the required properties of smoothness and exit at first-order exit points ends up dynamics that make no sense, developing undesirable singularities. Finally, Chapter 6 presents an extension of the moments method to co-dimension $3$ and higher: this is the first result which provides a unique admissible solution for this problem.

In this thesis we consider two great challenges in computer simulations of partial differential equations: multiscale data, varying over multiple scales in space and time, and data uncertainty, due to lack of or inexact measurements. We develop a multiscale method based on a coarse scale correction, using localized fine scale computations. We prove that the error in the solution produced by the multiscale method decays independently of the fine scale variation in the data or the computational domain. We consider the following aspects of multiscale methods: continuous and discontinuous underlying numerical methods, adaptivity, convection-diffusion problems, Petrov-Galerkin formulation, and complex geometries. For uncertainty quantification problems we consider the estimation of p-quantiles and failure probability. We use spatial a posteriori error estimates to develop and improve variance reduction techniques for Monte Carlo methods. We improve standard Monte Carlo methods for computing p-quantiles and multilevel Monte Carlo methods for computing failure probability.

This dissertation presents a sampling-based method that can be used for uncertainty quantification and deterministic or probabilistic optimization. The objective is to simultaneously address several difficulties faced by classical techniques based on response values and their gradients. In particular, this research addresses issues with discontinuous and binary (pass or fail) responses, and multiple failure modes. All methods in this research are developed with the aim of addressing problems that have limited data due to high cost of computation or experiment, e.g. vehicle crashworthiness, fluid-structure interaction etc.The core idea of this research is to construct an explicit boundary separating allowable and unallowable behaviors, based on classification information of responses instead of their actual values. As a result, the proposed method is naturally suited to handle discontinuities and binary states. A machine learning technique referred to as support vector machines (SVMs) is used to construct the explicit boundaries. SVM boundaries can be highly nonlinear, which allows one to use a single SVM for representing multiple failure modes.One of the major concerns in the design and uncertainty quantification communities is to reduce computational costs. To address this issue, several adaptive sampling methods have been developed as part of this dissertation. Specific sampling methods have been developed for reliability assessment, deterministic optimization, and reliability-based design optimization. Adaptive sampling allows the construction of accurate SVMs with limited samples. However, like any approximation method, construction of SVM is subject to errors. A new method to quantify the prediction error of SVMs, based on probabilistic support vector machines (PSVMs) is also developed. It is used to provide a relatively conservative probability of failure to mitigate some of the adverse effects of an inaccurate SVM. In the context of reliability assessment, the proposed method is presented for uncertainties represented by random variables as well as spatially varying random fields.In order to validate the developed methods, analytical problems with known solutions are used. In addition, the approach is applied to some application problems, such as structural impact and tolerance optimization, to demonstrate its strengths in the context of discontinuous responses and multiple failure modes.

In the present work we implemented parallel version of a computational fluid dynamics code. This code is based on Discontinuous Galerkin Method (DGM), which is due to its favourable properties suitable for parallelization. In the work we describe the Navier-Stokes equations and their discretization using DGM. We explain the advantages of usage of the DGM and formulate the serial algorithm. Next we focus on the parallel implementation of the algorithm and several particular issues connected to the parallelization. We present the numerical experiments showing the efficiency of the parallel code in the last chapter.

This thesis is concerned on the simulation of wet steam flow using discontinuous Galerkin method. Wet steam flow equations consist of Naviere-Stokes equations for compressible flow and Hill's equations for condensation of water vapor. The first part of this thesis describes the mathematical formulation of wet steam model and the derivation of Hill's equations. The model equations are discretized with the aid of discontinuous Galerkin method and backward difference formula which leads to implicit scheme represented by nonlinear algebraic system. This system is solved using Newton-like method. The derived scheme was implemented in program ADGFEM which is used for solving non-stationary convective-diffusive problems. The numerical results are presented in the last part of this thesis. 1

Neisseria meningitidis B libera vesículas de membrana externa, conhecidas pela sigla OMV. Essas possuem os mesmos componentes da membrana externa da bactéria e podem ser utilizadas como antígenos em vacinas contra a meningite B. As vesículas devem, também, expressar proteínas da membrana externa (OMP) e proteínas reguladoras do íon ferro (IRP). O objetivo deste trabalho é estudar a cinética de crescimento bacteriano, consumo das fontes de carbono e nitrogênio - especialmente os limitantes de crescimento ? e produção de OMV visando melhorar a produção desse antígeno. Realizaram-se cultivos descontínuos em biorreator, com duração de 20 h, empregando meio de Catlin com limitação de ferro e modificações nas concentrações de lactato, aminoácidos e glicerol. As condições do cultivo foram: 4,2 L de meio, temperatura de 36°C, pressão de 0,5 atm, vazão de ar 1 L/min, agitação entre 250-850 rpm, controle de oxigênio dissolvido em 10% de saturação. Constatou-se que o lactato é a principal fonte de carbono limitante, embora somente se tem a hipótese de que o glicerol age como protetor mecânico. O ácido L-glutâmico é a principal fonte de nitrogênio consumida durante o cultivo. As OMV começaram a ser liberadas quantitativamente no início da fase estacionária de crescimento. Sendo que a melhor condição para a produção de OMV, valor 162,3 mg/L, é aquela em que as concentrações iniciais de lactato e aminoácidos foram duplicadas, 15,00 g/L e 2,93 g/L respectivamente. Através da análise do padrão eletroforético, confirmou-se a presença das principais proteínas de superfície, inclusive das IRPs. A integridade da OMV foi constatada por microscopia eletrônica. Assim, o antígeno obtido mostra-se passível de utilização na composição de vacina anti-meningocócica. / Neisseria meningitidis B liberates outer membrane vesicles known by the abbreviation OMV. These vesicles have the same components of the outer membrane of the bacteria and may be used as antigens in vaccines against meningitis B. The vesicles must also express outer membrane proteins (OMP) and iron regulated proteins (IRP). The aim of this paper is to study bacterial growth kinetics, carbon and nitrogen sources consumption ? specially those which limit growth ? and OMV production, seeking to improve the production of this antigen. Discontinuous bioreactor cultivations were carried out for a period of 20 hours in Catlin medium with iron restriction and modifications in lactate, amino acid, and glycerol concentrations. Cultivation conditions were: 4,2 L of medium, temperature at 36ºC, 0,5 atm, air flow rate of 1 L/min, agitation between 250-850 rpm, and dissolved oxygen control at 10% of saturation. It was verified that lactate is the main limiting carbon source, although there is just a hypothesis that glycerol acts as a mechanic protector. The L-glutamic acid is the main source of nitrogen consumed during the cultivation. The OMV started to be liberated quantitatively at the beginning of the stationary phase of growth. The best condition for production of OMV, value 162,3 mg/L, is that where the initial concentrations of lactate and amino acids were duplicated, 15,00 g/L and 2,93 g/L, respectively. Through an analysis of the electroforetic pattern, the presence of the main surface proteins was confirmed, including the IRPs. The integrity of the OMV was testified by electronic microscopy. So, the antigen thus obtained may be used in the antimeningococcal vaccine composition.

L'objectif général de la thèse est la simulation numérique des mouvements forts du sol dûs aux séismes. Les déformations importantes du sol engendrent des comportements nonlinéaires dans les couches superficielles. L'apport principal de la thèse est la prise en compte de la nonlinéarité des milieux dans un contexte éléments finis Galerkin discontinus. Différentes lois de comportement sont implémentées et analysées. Le cas particulier du modèle élastoplastique de Masing-Prandtl-Ishlinskii-Iwan (MPII) est approfondi. Cette étude est divisée en deux parties. Une première qui vise à poser la structure du problème en présentant les équations et modèles utilisés pour décrire les mouvements du sol. Dans cette partie nous présentons également la méthode d’approximation spatiale Galerkin Discontinue ainsi que les différents schémas temporels que nous avons considérés. Une attention particulière est portée sur la complexité algorithmique du modèle nonlinéaire élastoplastique MPII en vue de réduire le temps de calcul des simulations. La deuxième partie est dédiée aux applications numériques. Ces applications sont réparties en trois catégories distinctes. 1) Nous nous intéressons toutd’abord à la configuration unidimensionnelle où une seule onde de cisaillement est propagée. Dans ce contexte, un flux numérique décentré est établi et des applications aux cas nonlinéaire élastique et nonlinéaire élastoplastique sont étudiées. Une solution analytique concernant le cas nonlinéaire élastique est proposée, ce qui permet de réaliser une étude numérique de convergence. 2) Le problème unidimensionnel étendu aux trois composantes du mouvement est étudié et utilisé comme un premier pas vers le 3D compte tenu du couplage entre les ondes de cisaillement et de compression. Nous nous intéressons ici à des signaux synthétiques et réels. L’application d’une méthode permettant de réduire significativement le temps de calcul du modèle élastoplastique est détaillée. 3) Une configuration tridimensionnelle est examinée. Après différentes applications de vérification en milieu linéaire, deux cas d’étude élastoplastique sont analysés. Une première sur un mode propre d’un cube puis une seconde sur un milieu plus réaliste composé d’un bassin hémisphérique à couches sédimentaires ayant un comportement élastoplastique / The general objective of this thesis is the numerical simulation of strong ground motions due to earthquakes. Significant deformations of the soil generate nonlinear behaviors in the superficial layers. The main contribution of this work is to take into account the nonlinearity of the media in a discontinuous Galerkin finite elements context. Different constitutive laws are implemented and analyzed. The particular case of theMasing-Prandtl-Ishlinskii-Iwan (MPII) elastoplastic model is looked at in-depth. This study is divided into two parts. A first one that aims at defining the framework of the problem by presenting the equations and models used to describe the soil motion. In this part we also present the Galerkin Discontinuous spatial approximation method as well as the different temporal schemes that we considered. Particular attention is paid to the algorithmic complexity of the nonlinear elastoplastic MPII model in order to reduce the computation time of simulations. The second part is dedicated to numerical applications. These applications are divided into three distinct categories. 1) We are first interested in the one-dimensional configuration where a single shear wave is propagated. In this context, an upwind numerical flux is established and applications to nonlinear elastic and nonlinear elastoplastic cases are studied. Ananalytical solution concerning the nonlinear elastic case is proposed, which makes it possible to carry out a numerical study of convergence. 2) The one-dimensional problem extended to the three components of the motion is studied and used as a first step towards 3D applications considering the coupling between the shear and compression waves. We are interested here in synthetic and real input signals. The application of a method that significantly reduces the calculation time of the elastoplastic model

Cette thèse propose l’étude d’une méthode numérique permettant de simuler un plasma. On considère un ensemble de particules, dont le mouvement est régi par l’équation de Vlasov, et qui est sensible aux forces électromagnétiques, qui proviennent des équations de Maxwell. La résolution numérique des équations de Vlasov-Maxwell est réalisée par une méthode Particle In Cell (PIC). La résolution des équations de Maxwell nécessite un maillage suffisamment fin afin de modéliser correctement les problémes multi-échelles que nous souhaitons traiter. Cependant, mailler finement tout le domaine de calcul a un coût. La nouveauté de cette thèse est de proposer un solveur PIC sur des maillages cartésiens localement raffinés, des maillages non conformes, afin de garantir la bonne modélisation du phénomène physique et d’éviter une trop forte pénalisation des temps de calcul.Nous utilisons une méthode Galerkin Discontinue en domaine temporelle (GDDT), qui offre l’avantage d’être d'une grande flexibilité dans le choix du maillage et qui est une méthode d’ordre élevé. Un point fondamental dans l’étude des solveurs PIC concerne le respect de la conservation de la charge. Nous proposons deux approches afin de traiter cet aspect. La première concerne les méthodes utilisant un système de Maxwell augmenté, dont la nouveauté a été de les étendre aux maillages non conformes. La seconde approche repose sur une méthode originale de pré-traitement du calcul du terme source de courant. / This thesis deals with the study of a numerical method to simulate a plasma. We consider a set of particles whose displacement is governed by the Vlasov equation and which creates an electromagnetic field thanks to Maxwell equations. The numerical resolution of the Vlasov-Maxwell system is performed by a Particle In Cell (PIC) method. The resolution of Maxwell equations needs a sufficiently fine mesh to correctly simulate the multi scaled problems that we have to face. Yet, a uniform fine mesh of the whole domain has a prohibitive cost. The novelty of this thesis is a PIC solver on locally refined Cartesian meshes : non conforming meshes, to guarantee the good modeling of the physical phenomena and to avoid too large CPU time. We use the Discontinuous Galerkin in Time Domain (DGTD) method which has the advantage of a great flexibility in the choice of the mesh and which is a high order method. A fundamental point in the study of PIC solvers is the respect of the charge conserving law. We propose two approaches to tackle this point. The first one deals with augmented Maxwell systems, that we have adapted to non conforming meshes. The second one deals with an original method of preprocessing of the calculation of the current source term.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O presente trabalho consiste em um estudo crítico do romance Sem tecto, entre ruínas, de Augusto Abelaira. Inicialmente é feita a análise sobre as possíveis consequências de uma apreensão descontínua e material do tempo, seus efeitos nas relações entre o protagonista e seu mundo, que se apresentam sob a forma de tédio e paralisia. Para a crítica proposta foram retomados conceitos descontinuidade e homogeneidade apresentados pelo filósofo Henri Bergson, além da noção de desligamento e desejo demonstrados pela sociologia de Zigmunt Bauman. O próximo alvo contemplado na pesquisa é a presença da memória e da lembrança indicados sob o ponto de vista contínuo do tempo, também analisados à luz da filosofia de Bergson, além do pensamento de Gilles Deleuze. A partir desse caminho investigativo é possível pensar a condição do protagonista como um confronto de si, uma experiência do tempo que traz uma revelação mediante a condição de ser tarde demais / This work provides a critical analysis on the novel Sem tecto, entre ruínas by Augusto Abelaira. In the first instance this study begans with the analysis of the possible consequences aroused by a material apprehension of time as well as its impact on relations between the protagonist and the world in which he lives, wich are presented by the feelings of boredom and paralysis. For the proposed work were taken up concepts of discontinuance and homogeneity created by the philosopher Henri Bergson, as well as the notion of detachment and desire demonstrated by the sociology of Zigmunt Bauman. The next target contemplated in the research is the presence of memory and recollection indicated under the point of view of continuous time, also examined by the philosophy of Bergson, in addition to the thought of Gilles Deleuze. Based on this investigative path is possible to consider the condition of the protagonist as a confront with itself, an experience of time that brings up a revelation upon the condition of being too late