The groundwater specialists and the reservoir engineers share the same interest in simulating multiphase flow in soil with heterogeneous intrinsic properties. They also both face the challenge of going from a well-modeled micrometer scale to the reservoir scale with a controlled loss of information. This upscaling process is indeed worthy to make simulation over an entire reservoir manageable and stochastically repeatable. Two upscaling steps can be defined: one from the micrometer scale to the Darcy scale, and another from the Darcy scale to the reservoir scale. In this thesis, a new second upscaling multiscale algorithm Finite Volume Mixed Hybrid Multiscale Methods (Fv-MHMM) is investigated. Extension to a two-phase flow system is done by weakly and sequentially coupling saturation and pressure via IMPES-like method.

The main aim of this thesis is to design efficient and novel numerical algorithms for a class of deterministic and stochastic dynamical systems with multiple time scales. Classical numerical methods for such problems need temporal resolution to resolve the finest scale and become, therefore, inefficient when the much longer time intervals are of interest. In order to accelerate computations and improve the long time accuracy of numerical schemes, we take advantage of various multiscale structures established from a separation of time scales. This dissertation is organized into four chapters: an introduction followed by three chapters, each based on one of three different papers. The framework of the heterogeneous multiscale method (HMM) is considered as a general strategy both for the design and the analysis of multiscale methods. In Chapter 2, we consider a new class of multiscale methods that use a technique related to the construction of a Poincaré map. The main idea is to construct effective paths in the state space whose projection onto the slow subspace shows the correct dynamics. More precisely, we trace the evolution of the invariant manifold M(t), identified by the level sets of slow variables, by introducing a slowly evolving effective path which crosses M(t). The path is locally constructed through interpolation of neighboring points generated from our developed map. This map is qualitatively similar to a Poincaré map, but its construction is based on the procedure which solves two split equations successively backward and forward in time only over a short period. This algorithm does not require an explicit form of any slow variables. In Chapter 3, we present efficient techniques for numerical averaging over the invariant torus defined by ergodic dynamical systems which may not be mixing. These techniques are necessary, for example, in the numerical approximation of the effective slow behavior of highly oscillatory ordinary differential equations in weak near-resonance. In this case, the torus is embedded in a higher dimensional space and is given implicitly as the intersection of level sets of some slow variables, e.g. action variables. In particular, a parametrization of the torus may not be available. Our method constructs an appropriate coordinate system on lifted copies of the torus and uses an iterated convolution with respect to one-dimensional averaging kernels. Non-uniform invariant measures are approximated using a discretization of the Frobenius-Perron operator. These two numerical averaging strategies play a central role in designing multiscale algorithms for dynamical systems, whose fast dynamics is restricted not to a circle, but to the tori. The efficiency of these methods is illustrated by numerical examples. In Chapter 4, we generalize the classical two-scale averaging theory to multiple time scale problems. When more than two time scales are considered, the effective behavior may be described by the new type of slow variables which do not have formally bounded derivatives. Therefore, it is necessary to develop a theory to understand them. Such theory should be applied in the design of multiscale algorithms. In this context, we develop an iterated averaging theory for highly oscillatory dynamical systems involving three separated time scales. The relevant multiscale algorithm is constructed as a family of multilevel solvers which resolve the different time scales and efficiently computes the effective behavior of the slowest time scale. / text

Numerous efforts have contributed to the study of phase-change problems for over a century|both analytical and numerical. Among those numerical approximations applied to solve phase-transition problems, phase-field models attract more and more attention because they not only capture two important effects, surface tension and supercooling, but also enable explicitly labeling the solid and liquid phases and the position of the interface. In the research of this dissertation, a phase-field model has been employed to simulate 2-D dendrite growth of pure nickel without a flow, and 2-D ice crystal growth in a high-Reynolds-number lid-driven-cavity flow. In order to obtain the details of ice crystal structures as well as the flow field behavior during freezing for the latter simulation, it is necessary to solve the phase-field model without convection and the equations of motion on two different scales. To accomplish this, a heterogeneous multiscale method is implemented for the phase-field model with convection such that the phase-field model is simulated on a microscopic scale and the equations of motion are solved on a macroscopic scale. Simulations of 2-D dendrite growth of pure nickel provide the validation of the phase-field model and the study of dendrite growth under different conditions, e.g., degree of supercooling, interface thickness, kinetic coefficient, and shape of the initial seed. In addition, simulations of freezing in a lid-driven-cavity flow indicate that the flow field has great effect on the small-scale dendrite structure and the flow eld behavior on the large scale is altered by freezing inside it.

In this dissertation, we develop multiscale finite element methods and uncertainty quantification technique for Richards' equation, a mathematical model to describe fluid flow in unsaturated porous media. Both coarse-level and fine-level numerical computation techniques are presented. To develop an accurate coarse-scale numerical method, we need to construct an effective multiscale map that is able to capture the multiscale features of the large-scale solution without resolving the small scale details. With a careful choice of the coarse spaces for multiscale finite element methods, we can significantly reduce errors. We introduce several methods to construct coarse spaces for multiscale finite element methods. A coarse space based on local spectral problems is also presented. The construction of coarse spaces begins with an initial choice of multiscale basis functions supported in coarse regions. These basis functions are complemented using weighted local spectral eigenfunctions. These newly constructed basis functions can capture the small scale features of the solution within a coarse-grid block and give us an accurate coarse-scale solution. However, it is expensive to compute the local basis functions for each parameter value for a nonlinear equation. To overcome this difficulty, local reduced basis method is discussed, which provides smaller dimension spaces with which to compute the basis functions. Robust solution techniques for Richards' equation at a fine scale are discussed. We construct iterative solvers for Richards' equation, whose number of iterations is independent of the contrast. We employ two-level domain decomposition pre-conditioners to solve linear systems arising in approximation of problems with high contrast. We show that, by using the local spectral coarse space for the preconditioners, the number of iterations for these solvers is independent of the physical properties of the media. Several numerical experiments are given to support the theoretical results. Last, we present numerical methods for uncertainty quantification applications for Richards' equation. Numerical methods combined with stochastic solution techniques are proposed to sample conductivities of porous media given in integrated data. Our proposed algorithm is based on upscaling techniques and the Markov chain Monte Carlo method. Sampling results are presented to prove the efficiency and accuracy of our algorithm.

O presente trabalho trata da modelagem da condução de calor transiente com geração de calor em meios heterogêneos, e tem o objetivo de desenvolver um modelo multiescala adequado a esse fenômeno. Já existem modelos multiescala na literatura relacionados ao problema proposto, e que são válidos para os seguintes casos: (a) o elemento de volume representativo tem tamanho desprezível quando comparado ao comprimento característico macroscópico (e como consequência, a microescala tem inércia térmica desprezível); ou (b) a geração de calor é homogênea na microescala. Por outro lado, o modelo proposto nesta tese, o qual é desenvolvido utilizando uma descrição variacional do problema, pode ser aplicado a elementos de volume representativos finitos e em condições em que a geração de calor é heterogênea na microescala. A discretização temporal (diferenças finitas) e as discretizações espaciais na microescala e na macroescala (método dos elementos finitos) são apresentadas em detalhes, juntamente com os algoritmos necessários para implementar a solução do problema. Nesta tese são apresentados casos numéricos simples, procurando verificar não só o modelo teórico multiescala desenvolvido, mas também a implementação feita. Para tanto, são analisados, por exemplo, (a) casos em que considera-se a microescala um material homogêneo, tornando possível a comparação da solução multiescala com a solução convencional (uma única escala) pelo método dos elementos finitos, e (b) um caso em um material heterogêneo para o qual a solução completa, isto é, modelando diretamente os constituintes no corpo macroscópico, é obtida, tornando possível a comparação com a solução multiescala. A solução na microescala para vários casos analisados nesta tese sofre grande influência da inércia térmica da microescala. Para demonstrar o potencial de aplicação do modelo multiescala, simula-se a cura de um elastômero carregado com negro de fumo. Embora a simulação demonstre que a inércia térmica não precise ser considerada para esse caso em particular, a aplicação da presente metodologia torna possível modelar a cura do elastômero diretamente sobre a microescala, uma abordagem até então não utilizada no contexto de métodos multiescala. Essa metodologia abre a possibilidade para futuros aperfeiçoamentos da modelagem do estado de cura. / This work deals with the modeling of transient heat conduction with heat generation in heterogeneous media, and its objective is to develop a proper multiscale model for this phenomenon. There already exist multiscale models in the literature related to this proposed problem, and which are valid for the following cases: (a) the representative volume element has a negligible size when compared to the characteristic macroscopic size (and, as a consequence, the microscale has a negligible thermal inertia); or (b) the heat generation is homogeneous at the microscale. On the other hand, the model proposed in this thesis, which is developed using a variational description of the problem, can be applied to finite representative volume elements and in conditions in which the heat generation is heterogeneous at the microscale. The time discretization (finite difference) and the space discretizations at both the microscale and the macroscale (finite element method) are presented in details, together with the algorithms needed for implementing the solution of the problem. In this thesis, simple numerical cases are presented, aiming to verify not only the theoretical multiscale model developed, but also its implementation. For this, it is analyzed, for instance, (a) cases in which the microscale is taken as a homogeneous material, making it possible the comparison of the multiscale solution with the conventional solution (one single scale) by the finite element method, and (b) a case in a heterogeneous material for which the full solution, that is, modeling all constituents directly on the macroscale, is obtained, making it possible the comparison with the multiscale solution. The solution at the microscale for several cases analyzed in this thesis suffers a large influence of the microscale thermal inertia. To demonstrate the application potential of the multiscale model, the cure of a carbon black loaded elastomer is simulated. Although the simulation shows that the thermal inertia does not have to be considered for this case in particular, the application of the present methodology makes it possible to model the cure of the elastomer directly at the microscale, an approach not used in multiscale methods context until now. This methodology opens the possibility for future improvements of the state of cure modeling.

O presente trabalho trata da modelagem da condução de calor transiente com geração de calor em meios heterogêneos, e tem o objetivo de desenvolver um modelo multiescala adequado a esse fenômeno. Já existem modelos multiescala na literatura relacionados ao problema proposto, e que são válidos para os seguintes casos: (a) o elemento de volume representativo tem tamanho desprezível quando comparado ao comprimento característico macroscópico (e como consequência, a microescala tem inércia térmica desprezível); ou (b) a geração de calor é homogênea na microescala. Por outro lado, o modelo proposto nesta tese, o qual é desenvolvido utilizando uma descrição variacional do problema, pode ser aplicado a elementos de volume representativos finitos e em condições em que a geração de calor é heterogênea na microescala. A discretização temporal (diferenças finitas) e as discretizações espaciais na microescala e na macroescala (método dos elementos finitos) são apresentadas em detalhes, juntamente com os algoritmos necessários para implementar a solução do problema. Nesta tese são apresentados casos numéricos simples, procurando verificar não só o modelo teórico multiescala desenvolvido, mas também a implementação feita. Para tanto, são analisados, por exemplo, (a) casos em que considera-se a microescala um material homogêneo, tornando possível a comparação da solução multiescala com a solução convencional (uma única escala) pelo método dos elementos finitos, e (b) um caso em um material heterogêneo para o qual a solução completa, isto é, modelando diretamente os constituintes no corpo macroscópico, é obtida, tornando possível a comparação com a solução multiescala. A solução na microescala para vários casos analisados nesta tese sofre grande influência da inércia térmica da microescala. Para demonstrar o potencial de aplicação do modelo multiescala, simula-se a cura de um elastômero carregado com negro de fumo. Embora a simulação demonstre que a inércia térmica não precise ser considerada para esse caso em particular, a aplicação da presente metodologia torna possível modelar a cura do elastômero diretamente sobre a microescala, uma abordagem até então não utilizada no contexto de métodos multiescala. Essa metodologia abre a possibilidade para futuros aperfeiçoamentos da modelagem do estado de cura. / This work deals with the modeling of transient heat conduction with heat generation in heterogeneous media, and its objective is to develop a proper multiscale model for this phenomenon. There already exist multiscale models in the literature related to this proposed problem, and which are valid for the following cases: (a) the representative volume element has a negligible size when compared to the characteristic macroscopic size (and, as a consequence, the microscale has a negligible thermal inertia); or (b) the heat generation is homogeneous at the microscale. On the other hand, the model proposed in this thesis, which is developed using a variational description of the problem, can be applied to finite representative volume elements and in conditions in which the heat generation is heterogeneous at the microscale. The time discretization (finite difference) and the space discretizations at both the microscale and the macroscale (finite element method) are presented in details, together with the algorithms needed for implementing the solution of the problem. In this thesis, simple numerical cases are presented, aiming to verify not only the theoretical multiscale model developed, but also its implementation. For this, it is analyzed, for instance, (a) cases in which the microscale is taken as a homogeneous material, making it possible the comparison of the multiscale solution with the conventional solution (one single scale) by the finite element method, and (b) a case in a heterogeneous material for which the full solution, that is, modeling all constituents directly on the macroscale, is obtained, making it possible the comparison with the multiscale solution. The solution at the microscale for several cases analyzed in this thesis suffers a large influence of the microscale thermal inertia. To demonstrate the application potential of the multiscale model, the cure of a carbon black loaded elastomer is simulated. Although the simulation shows that the thermal inertia does not have to be considered for this case in particular, the application of the present methodology makes it possible to model the cure of the elastomer directly at the microscale, an approach not used in multiscale methods context until now. This methodology opens the possibility for future improvements of the state of cure modeling.

O presente trabalho trata da modelagem da condução de calor transiente com geração de calor em meios heterogêneos, e tem o objetivo de desenvolver um modelo multiescala adequado a esse fenômeno. Já existem modelos multiescala na literatura relacionados ao problema proposto, e que são válidos para os seguintes casos: (a) o elemento de volume representativo tem tamanho desprezível quando comparado ao comprimento característico macroscópico (e como consequência, a microescala tem inércia térmica desprezível); ou (b) a geração de calor é homogênea na microescala. Por outro lado, o modelo proposto nesta tese, o qual é desenvolvido utilizando uma descrição variacional do problema, pode ser aplicado a elementos de volume representativos finitos e em condições em que a geração de calor é heterogênea na microescala. A discretização temporal (diferenças finitas) e as discretizações espaciais na microescala e na macroescala (método dos elementos finitos) são apresentadas em detalhes, juntamente com os algoritmos necessários para implementar a solução do problema. Nesta tese são apresentados casos numéricos simples, procurando verificar não só o modelo teórico multiescala desenvolvido, mas também a implementação feita. Para tanto, são analisados, por exemplo, (a) casos em que considera-se a microescala um material homogêneo, tornando possível a comparação da solução multiescala com a solução convencional (uma única escala) pelo método dos elementos finitos, e (b) um caso em um material heterogêneo para o qual a solução completa, isto é, modelando diretamente os constituintes no corpo macroscópico, é obtida, tornando possível a comparação com a solução multiescala. A solução na microescala para vários casos analisados nesta tese sofre grande influência da inércia térmica da microescala. Para demonstrar o potencial de aplicação do modelo multiescala, simula-se a cura de um elastômero carregado com negro de fumo. Embora a simulação demonstre que a inércia térmica não precise ser considerada para esse caso em particular, a aplicação da presente metodologia torna possível modelar a cura do elastômero diretamente sobre a microescala, uma abordagem até então não utilizada no contexto de métodos multiescala. Essa metodologia abre a possibilidade para futuros aperfeiçoamentos da modelagem do estado de cura. / This work deals with the modeling of transient heat conduction with heat generation in heterogeneous media, and its objective is to develop a proper multiscale model for this phenomenon. There already exist multiscale models in the literature related to this proposed problem, and which are valid for the following cases: (a) the representative volume element has a negligible size when compared to the characteristic macroscopic size (and, as a consequence, the microscale has a negligible thermal inertia); or (b) the heat generation is homogeneous at the microscale. On the other hand, the model proposed in this thesis, which is developed using a variational description of the problem, can be applied to finite representative volume elements and in conditions in which the heat generation is heterogeneous at the microscale. The time discretization (finite difference) and the space discretizations at both the microscale and the macroscale (finite element method) are presented in details, together with the algorithms needed for implementing the solution of the problem. In this thesis, simple numerical cases are presented, aiming to verify not only the theoretical multiscale model developed, but also its implementation. For this, it is analyzed, for instance, (a) cases in which the microscale is taken as a homogeneous material, making it possible the comparison of the multiscale solution with the conventional solution (one single scale) by the finite element method, and (b) a case in a heterogeneous material for which the full solution, that is, modeling all constituents directly on the macroscale, is obtained, making it possible the comparison with the multiscale solution. The solution at the microscale for several cases analyzed in this thesis suffers a large influence of the microscale thermal inertia. To demonstrate the application potential of the multiscale model, the cure of a carbon black loaded elastomer is simulated. Although the simulation shows that the thermal inertia does not have to be considered for this case in particular, the application of the present methodology makes it possible to model the cure of the elastomer directly at the microscale, an approach not used in multiscale methods context until now. This methodology opens the possibility for future improvements of the state of cure modeling.

L’étude des écoulements souterrains et l’ingénierie réservoir partagent le même intérêt pour la simulation d’écoulement multiphasique dans des sols aux propriétés intrinsèquement hétérogènes. Elles rencontrent également les mêmes défis pour construire un modèle à l’échelle réservoir en partant de données micrométriques tout en contrôlant la perte d’informations. Ce procédé d’upscaling est utile pour rendre les simulations faisables et répétables dans un cadre stochastique. Deux processus de mise à l’échelle sont définis: l’un depuis l’échelle micrométrique jusqu’à l’échelle de Darcy, et, un autre depuis l’échelle de Darcy vers l’échelle du réservoir. Dans cette thèse, un nouvel algorithme traitant du second upscaling Finite Volume Mixed Hybrid Multiscale Method (FV-MHMM) est étudié. L’extension au diphasique est faite au moyen d’un couplage séquentiel faible entre saturation et pression grâce à une méthode de type IMPES. / The groundwater specialists and the reservoir engineers share the same interest in simulating multiphase flow in soil with heterogeneous intrinsic properties. They also both face the challenge of going from a well-modeled micrometer scale to the reservoir scale with a controlled loss of information. This upscaling process is indeed worthy to make simulation over an entire reservoir manageable and stochastically repeatable. Two upscaling steps can be defined: one from the micrometer scale to the Darcy scale, and another from the Darcy scale to the reservoir scale. In this thesis, a new second upscaling multiscale algorithm Finite Volume Mixed Hybrid Multiscale Methods (Fv-MHMM) is investigated. Extension to a two-phase flow system is done by weakly and sequentially coupling saturation and pressure via IMPES-like method.

Ce travail concerne principalement la détermination des propriétés acoustiques de mousses. Il s’agit d’un projet mené dans le cadre d’une collaboration entre une équipe de physico-chimie des mousses chargée de l’élaboration de matériaux modèles (laboratoire Navier UMR 8205 CNRS) et une équipe d’acousticiens chargée de l’étude de leurs propriétés acoustiques (laboratoire MSME UMR 8208 CNRS). Cette thèse s’articule essentiellement autour de trois parties principales, dont le contenu est résumé ci-dessous. 1) La première partie porte sur la génération de surfaces de réponse par des approximations polynomiales, dans le but de disposer d'un modèle intermédiaire entre le modèle éléments finis micro-macro et la réponse macroscopique. Au lieu d'appeler le modèle éléments finis systématiquement dans un travail d'optimisation, on a recourt à la surface de réponse qui contient l'information associée aux points de calcul éléments finis ainsi que les interpolations correspondantes. Ce manuscrit a été publié dans le journal AAA sous forme de communication rapide. 2) La deuxième partie porte sur la mise au point d'un modèle semi-analytique définit à partir d'une formule disponible pour prédire la perméabilité d'une plaque infinie percée par un trou de surface connue. Ce modèle, utilisé de manière appropriée, permet de calculer la perméabilité de mousses dont la taille de bulles est constante et le taux de fermeture de membranes variable. Des validations numériques par éléments finis et expérimentales sont proposées. L'article a été accepté pour publication dans la revue Physical Review E. 3) La troisième partie, porte sur un calcul éléments finis dans lequel un grand nombre de réalisations sont menées de manière à prendre en compte l'ensemble des combinaisons possibles lorsque on dispose de caractérisation expérimentales fines à l'échelle de la microstructure et que l'on souhaite connaitre la réponse de la mousse avec précision. Le manuscrit est en préparation et la revue visée pour ce dernier manuscrit est le journal Materials and Design. Une introduction et une conclusion générale complètent ces trois parties, et permettent de mettre en perspectives ces contributions par rapport à la littérature existante sur le sujet / This work mainly concerns the determination of the acoustic properties of foams. This is a project carried out as part of a collaboration between a team of physico-chemistry of foams in charge of the development of model materials (Navier laboratory UMR 8205 CNRS) and a team of acousticians responsible for the study of their acoustic properties (MSME laboratory UMR 8208 CNRS). This thesis is structured around three main parts, the content of which is summarized below. 1) The first part deals with the generation of response surfaces by polynomial approximations, in order to have an intermediate model between the micro-macro finite element model and the macroscopic response. Instead of calling the finite element model systematically in an optimization work, we use the response surface that contains the information associated with finite element calculation points and the corresponding interpolations. This manuscript was published in the AAA journal as a fast track publication. 2) The second part focuses on the development of a semi-analytical model defined from an available formula to predict the permeability of a circular orifice in a thin plate. This model, used in an appropriate way, makes it possible to calculate the permeability of foams with a constant bubble size but a tuned membrane content. Numerical validations by finite element computations are proposed. The article has been accepted for publication in the journal Physical Review E. 3) The third part deals with a finite element calculation in which a large number of realizations are carried out in order to take into account all the possible combinations when one has fine experimental characterization at the microstructure scale and that one seek to determine the properties of the foam with precision. The manuscript is in preparation and a possible journal for the publication of this manuscript is the journal Materials and Design. An introduction and a general conclusion complete these three parts, and make it possible to discuss these contributions

Cette thèse se concentre sur le développement d'une méthode précise et efficace pour la simulation des grandes échelles (LES) des écoulements turbulents. Une approche de la LES basée sur la méthode variationnelle multi-échelles (VMS) est considérée. La VMS applique aux équations de la dynamique des fluides une séparation d'échelles a priori sans recours à des hypothèses sur les conditions aux limites ou sur l'uniformité du maillage. Afin d'assurer effectivement une séparation d'échelles dans l'espace des nombres d'onde associé, nous choisissons d'utiliser les ondelettes de deuxième génération (SGW), une base polynomiale qui présente des propriétés de localisation spatiale-fréquence optimales. A partir de la séparation d'échelles ainsi réalisée, l'action du modèle sous-maille est limitée à un intervalle de nombres d'onde proche de la coupure spectrale. Cette approche VMS-LES basée sur les ondelettes est désignée par WAVVMS-LES. Elle est incorporée dans un solveur d'ordre élevé pour la simulation des écoulements incompressibles sur la base d'une méthode de Galerkin discontinue (DG-FEM) stabilisée pour la pression. La méthode est évaluée par réalisation de LES sur des maillages fortement sous-résolus pour le cas test du tourbillon de Taylor-Green 3D à deux nombres de Reynolds différents. / This thesis focuses on the development of an accurate and efficient method for performing Large-Eddy Simulation (LES) of turbulent flows. An LES approach based upon the Variational Multiscale (VMS) method is considered. VMS produces an a priori scale-separation of the governing equations, in a manner which makes no assumptions on the boundary conditions and mesh uniformity. In order to ensure that scale-separation in wavenumber is achieved, we have chosen to make use of the Second Generation Wavelets (SGW), a polynomial basis which exhibits optimal space-frequency localisation properties. Once scale-separation has been achieved, the action of the subgrid model is restricted to the wavenumber band closest to the cutoff. We call this approach wavelet-based VMS-LES (WAV-VMS-LES). This approach has been incorporated within the framework of a high-order incompressible flow solver based upon pressure-stabilised discontinuous Galerkin FEM (DG-FEM). The method has been assessed by performing highly under-resolved LES upon the 3D Taylor-Green Vortex test case at two different Reynolds numbers.