L'étude de la susceptibilité électromagnétique de circuits électroniques nécessite l'utilisation d'un outil de simulation rapide, précis et suffisamment flexible pour intégrer les dernières innovations technologiques. La méthode itérative basée sur le concept d'onde (notée WCIP pour Wave Concept Iterative Procedure) initialement proposée par H. Baudrand est particulièrement adaptée pour la modélisation numérique de circuits multicouches à plusieurs niveaux de métallisation. Pour ce type de circuits, elle se révèle être l'une des méthodes qui utilise le plus petit nombre d'inconnues pour atteindre une précision donnée. Néanmoins, la WCIP n'est pas adaptée à la prise en compte des diélectriques inhomogènes et des trous d'interconnexion. L'objectif de la thèse est de s'affranchir de ces limitations par un couplage avec des méthodes numériques volumiques. En premier lieu, l'hybridation a été mise en œuvre avec une méthode basée sur la théorie des lignes de transmission pour des raisons de correspondance de maillages. Par la suite, le couplage avec une technique d'éléments finis de type Galerkin Discontinu (notée GD) Hybridée permet d'atteindre des objectifs de précision et de rapidité car GD apporte une flexibilité dans la discrétisation. En effet, c'est une méthode d'éléments finis non conforme qui permet notamment de faire varier d'un élément à l'autre l'ordre polynomial d'approximation. On a ainsi développé une nouvelle méthode numérique hybride couplant la WCIP avec des méthodes volumiques qui offrent plus de souplesse pour la prise en compte des milieux complexes. Enfin, une stratégie de résolution par décomposition de domaines est également abordée à la fin du manuscrit. / Electromagnetic susceptibility study of electronic circuits requires the use of a simulation tool which is fast, accurate and flexible enough to incorporate the latest technological innovations. The Wave Concept Iterative Procedure (WCIP) initially proposed by H. Baudrand is particularly adapted for numerical modeling of multilayered circuits with multilevel metallization. For this kind of circuits, it turns out to be one of the methods that uses the smallest number of unknowns to reach a given accuracy. However, the WCIP is not appropriate for inhomogeneous dielectric substrates and metallized via holes. The aim of this PhD thesis is to overcome these limitations coupling the WCIP with volume numerical methods. First, hybridization is carried out with the Frequency Domain Transmission Line Matrix (denoted FDTLM) assuming matching meshes at the interface between computational domains of both methods. Subsequently, the coupling with a finite element technique like a Hybridized Discontinuous Galerkin (denoted DG) method is considered to achieve the objectives of accuracy and speed because DG brings flexibility in the discretization. Indeed, it is a nonconforming finite element method which allows in particular changing the polynomial approximation order from one element to another. Therefore, a new hybrid method is developed coupling the WCIP with volume numerical methods which offer more flexibility for dealing with complex environments. Finally, a domain decomposition solution strategy is also discussed at the end of the manuscript.

Studies on the phenomenon of hyperbaton (or discontinuous noun phrases) in Latin have traditionally suffered from over-generalizing descriptions and categorizations drawing on examples from a variety of sources. This thesis attempts a different approach, closely examining the totality of instances in a single text, Cicero’s Pro Milone, in view of establishing a preliminary typology and to investigate the motivations behind it. It also consistently compares the discontinuous data with continuous examples, in order to rule out as many alternative motivations as possible. The results suggest that the discursive value of the components is crucial for whether hyperbaton occurs or not: in particular a combination of a pragmatic function (Topic or Focus) and Contrast, and connects this to some recent theoretical proposals.

Made available in DSpace on 2014-06-11T19:32:50Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-03-18Bitstream added on 2014-06-13T18:07:15Z : No. of bitstreams: 1 cardin_pt_dr_sjrp.pdf: 479456 bytes, checksum: 52785d20631e0d11a14a241fde1ae7c9 (MD5) / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / Neste trabalho estudamos sistemas diferenciais forçados, também conhecidos como sistemas de equações com impasse. Estudamos os casos onde tais sistemas são suaves e os casos onde são possivelmente descontínuos. Usando técnicas de perturbação singular obtemos alguns resultados sobre a dinâmica destes sistemas em vizinhanças dos conjuntos de impasse. No caso suave, a Teoria de Fenichel clássica e crucial para o desenvolvimento dos principais resultados. Para o caso com descontinuidades, uma teoria similar a Teoria de Fenichel ´e desenvolvida. Além disso, estudamos a bifurcação de ciclos limites das órbitas periódicas de um centro diferencial linear quando perturbamos tal centro dentro de uma classe de sistemas diferenciais lineares por partes com impasse / In this work we study constrained differential systems, also known as systems of equations with impasse. We study the cases where such systems are smo oth and the cases where they are p ossibly discontinuous. Using singular p erturbation techniques we obtain some results on the dynamic of these systems in neighb orho o ds of the impasse sets. In smo oth case, the classical Fenichel’s Theory is crucial for the development of the main results. For the case with discontinuity, a similar theory to Fenichel’s Theory is develop ed. Moreover, we study the bifurcation of limit cycles from the p erio dic orbits of a linear differential center when we p erturb such center inside a class of piecewise linear differential systems with impasse

Cette thèse se penche sur la propagation d’ondes au sein du système coupléTerre-océan-atmosphère. La compréhension de ces phénomènes a une importance majeure pour l’étude de perturbations sismiques et d’explosions atmosphériques notamment dans le cadre de missions spatiales planétaires. Les formes d’ondes issues du couplage fluide-solide permettent d’obtenir de précieuses informations sur la source du signal ou les propriétés des milieux de propagation. On développe donc deux outils numériques d’ordre élevé pour la propagation d’ondes acoustiques et de gravité. L'u en différences finies et se concentre sur le milieu atmosphérique et la propagation d’ondes linéaires dans un milieu stratifié visqueux et avec du vent. Cette méthode linéaire est validée par des solutions quasi-analytiques reposant sur les équations de dispersion dans une atmosphère stratifiée. Elle est aussi appliquée à deux cas d’études : la propagation d’ondes liée à l’impact d’une météorite à la surface de Mars (mission NASA INSIGHT), et la propagation d’ondes atmosphériques liées au tsunami de Sumatra en 2004. La seconde méthode résout la propagation non-linéaire d’ondes gravito-acoustiques dans une atmosphère couplée, avec topographie, à la propagation d’ondes élastiques dans un solide visco-élastique. Cette méthode repose sur sur le couplage d’une formulation en éléments finis discontinus, pour résoudre les équations de Navier-Stokes la partie fluide, par éléments finis continus pour résoudre les équations de l’élastodynamique dans la partie solide. Elle a été validée grâce à des solutions analytiques ainsi que par des comparaisons avec les résultats de la méthode par différences finies. / This thesis deals with the wave propagation problem within the Earth-ocean-atmosphere coupled system. A good understanding of the these phenomena has a major importance for seismic and atmospheric explosion studies, especially for planetary missions. Atmospheric wave-forms generated by explosions or surface oscillations can bring valuable information about the source mechanism or the properties of the various propagation media. We develop two new numerical full-wave high-order modeling tools to model the propagation of acoustic and gravity waves in realistic atmospheres. The first one relies on a high-order staggered finite difference method and focus only on the atmosphere. It enables the simultaneous propagation of linear acoustic and gravity waves in stratified viscous and windy atmosphere. This method is validated against quasi-analytical solutions based on the dispersion equations for a stratified atmosphere. It has also been employed to investigate two cases : the atmospheric propagation generated by a meteor impact on Mars for the INSIGHT NASA mission and for the study of tsunami-induced acoutic and gravity waves following the 2004 Sumatra tsunami. The second numerical method resolves the non-linear acoustic and gravity wave propagation in a realistic atmosphere coupled, with topography, to the elastic wave propagation in a visco-elastic solid. This numerical tool relies on a discontinuous Galerkin method to solve the full Navier-Stokes equations in the fluid domain and a continuous Galerkin method to solve the elastodynamics equations in the solid domain. It is validated against analytical solutions and numerical results provided by the finite-difference method.

A prestudy indicated superior accuracy and convergence properties of apseudo-spectral method compared to a spline-based method implemented byCòrdoba et al. in 2005 when solving the α-patches problem. In this thesis wefurther investigate the numerical properties of the pseudo-spectral method and makeit more robust by implementing the Nonequispaced Fast Fourier Transform. Wepresent a more detailed overview and analysis of the pseudo-spectral method and theα-patches problem in general and conclude that the pseudo-spectral method issuperior in regards to accuracy in periodic settings.

O presente trabalho foi realizado para estudar o efeito sobre a propagação de fraturas em juntas não-persistentes quando existe uma variação da rugosidade nas juntas. Para esta abordagem, foram analisados os modos de coalescência, o ângulo de início das fraturas, a resistência à compressão e deformação de modelos de argamassa contendo juntas não-persistentes e submetidos a estados biaxiais de tensão. Na literatura especializada encontram-se vários trabalhos que estudam o comportamento de juntas lisas não-persistentes. Infelizmente, este tipo de juntas não leva em conta os efeitos da dilatância e do aumento do atrito. Neste estudo, desenvolveu-se um método para produzir juntas rugosas não-persistentes dentro dos moldes de argamassa. Desta forma, conseguiu-se comparar os diferentes modos de ruptura, tanto para as juntas lisas como as rugosas. A configuração geométrica usada foi de 15 juntas com 'alfa' = 54 graus; 'beta' = 60 graus; Lj = Lb = 50mm e d = 25mm. Onde 'alfa' é o ângulo gerado entre o plano de uma junta e o plano formado entre as pontas de juntas não-coplanares, 'beta' é a inclinação da junta com respeito ao plano de tensão maior, Lb denota à distância entre juntas paralelas coplanares, Lj é o comprimento da junta e d é a distância entre juntas paralelas não coplanares. Estes parâmetros mantiveram-se constantes em todos os ensaios, tanto para as juntas lisas (JRC = 0) como as rugosas (JRC = 6,96 e JRC = 12,25). Usando os parâmetros acima mencionados, observou-se que o tipo de ruptura sempre aconteceu por escalonamento. Para as amostras contendo as juntas lisas, JRC = 0, o mecanismo principal de coalescência entre as juntas é a tração. Para estas juntas, os ângulos de início das fraturas, em média, resultaram de 64º e 65º para esquerda e direita, respectivamente. Em média, a resistência a compressão normalizada ('sigma'nor = 'sigma'1 - 'sigma'2 / 'sigma'cs; onde 'sigma'1 e 'sigma'2 são as tensões principais e 'sigma'cs é a resistência média à compressão simples) destes corpos de prova resultou ter o valor mais baixo, 'sigma'nor = 0,52, e a maior deformação, 'épsilon' = 0,0057. No caso das juntas com JRC = 6,7, a coalescência ocorreu tanto por tração como por cisalhamento, descrevendo um caminho ondulante entre as pontas das juntas. O ângulo médio de início das fraturas foi de 40º para o lado esquerdo e 48º para o lado direito. A resistência média normalizada destes corpos a compressão foi de 'sigma'nor = 0,54 e uma deformação de 'épsilon' = 0,0053. Com as juntas com JRC = 12,3, a coalescência também apresentou ambos os mecanismos, tração e cisalhamento, seguindo uma direção inclinada no início e trajetória reta no meio. Os ângulos de início obtidos foram 5º no lado esquerdo e 20º do lado direito. Os valores maiores de resistência média normalizada e o valor médio menor de deformação foram obtidos nestes tipo de juntas, sendo estes de 'sigma'nor = 0,59 e 'épsilon' = 0,0045. Verificou-se então que a rugosidade tem uma grande influência sobre a propagação da fratura, afetando grandemente o modo de coalescência, o ângulo de início de propagação das fraturas e a resistência e deformação total dos corpos de prova / The present experimental study was conducted to investigate the effect of joint roughness on the fracture propagation of models with non persistent joints. For this approach the parameters investigated are : coalescence, the crack initiation angle, the compressive strength and deformation of the mortar samples containing non-persistent rough joints under biaxial loading. There are many works in specialized literature that study the behavior of non-persistent smooth joints, unfortunately, this kind of approach does not take into account the effect of the dilation and the increase of friction due to the joint roughness. In this study, a new method was developed to produce non-persistent rough joints inside the mortar models. With this method it was possible to compare the different paths of rupture generated for the smooth and for the rough joints. Each sample had 15 joints with 'alfa' = 54º; 'beta' = 60º; Lj = Lb = 50mm and d = 25mm. Where 'alfa' is the formed angle between the joint plane and the plane generated by two non coplanar joint tips, 'beta' is the joint inclination angle with the principal plane stress, Lb is the distance between coplanar joints, Lj is the length of the joint, d is the distance between two non-coplanar joints. These parameters remained constant in all the tests; the only variation permitted was in the joint roughness : from smooth joints (JRC = 0) to rough joints (JRC = 6,96 and JRC = 12,25). Using the mentioned parameters before it was always obtained the stepping failure. For the samples containing smooth joints, JRC = 0, the main mechanism of coalescence is tension. For these joints the average crack initiation angles, had resulted of 64º and 65º for left and right side, respectively. The average normalized compression strength ('sigma'nor = 'sigma'1 - 'sigma'2 / 'sigma'cs; where 'sigma'1 and 'sigma'2 are the principal stresses and 'sigma'cs is the average compression strength) of these tests resulted to have the lowest value, 'sigma'nor = 0,52, and the highest deformation, 'épsilon' = 0,0057. In the case of joints with JRC = 6,7; the mechanisms of coalescence are tension and shear, growing in a waving path between the joint tips. The average crack initiation angle was of 40º for the left side and 48º for the right side of the joint. The average normalized compression strength of these tests was of 'sigma'nor = 0,54 and deformation 'épsilon' = 0,0053. With joints having JRC = 12,3; the coalescence also presents both mechanisms, tension and shear, following a direction inclined in the beginning and a straight line in the middle of the way. The crack initiation angles had been : 5º in the left side and 20º of the right side. The highest values of average normalized strength resistant and the lowest average value of deformation were found in this type of joint, being of 'sigma'nor = 0,59 e 'épsilon' = 0,0045 respectively. With this approach, it was verified that the joint roughness influence the fracture propagation, affecting the coalescence, the crack initiation angle, the resistance and total deformation of the tested specimens

A pesquisa desenvolve o projeto e a implementaÃÃo de um microinversor monofÃsico de estÃgio Ãnico obtido da derivaÃÃo de quatro conversores flybacks CC-CC operando no modo descontÃnuo e associado em mÃdulos paralelos. Este arranjo reduz o volume dos magnÃticos, diminui os esforÃos de correntes nos primÃrios dos transformadores, bem como minimiza as oscilaÃÃes no mÃdulo fotovoltaico. Em meio as principais propriedades do microinversor flyback (MIF) desta pesquisa, as chaves dos primÃrios atuam com comutaÃÃo SPWM em alta frequÃncia e defasadas em cento e oitenta graus, enquanto as chaves dos secundÃrios sÃo comandadas por PWMâs complementares. Para validar a pesquisa serà desenvolvido um protÃtipo que efetuarà o processamento de energia interligando um painel fotovoltaico com 200 W de potÃncia, em cargas resistivas com caracterÃsticas senoidais. O sistema proposto està fundamentado na configuraÃÃo mÃdulo inversor integrado (MIC), que por sua vez, possui menores perdas nos semicondutores e elementos magnÃticos da topologia, quando comparado aos complexos sistemas fotovoltaicos centralizados. Dentre as aplicaÃÃes do microinversor flyback encontram-se: abastecimento de sistemas remotos, alimentaÃÃo de aparelhos eletroeletrÃnicos, pequenas estaÃÃes de bombeamentos de Ãgua, conexÃes a rede elÃtrica, entre outros. Apesar da baixa potÃncia, o microinversor apresenta caracterÃsticas elÃtricas essenciais a inserÃÃo na matriz energÃtica, tais como: isolaÃÃo entre a fonte e a carga, formas de ondas de tensÃes e correntes na saÃda com pequenas distorÃÃes e na frequÃncia da rede. Os resultados obtidos em laboratÃrio revelam que o microinversor flyback operando em malha aberta obtÃm rendimentos prÃximos a 90%, sendo entregue à carga cerca de 180 W. Sendo assim, a facilidade na construÃÃo, a utilizaÃÃo de poucos componentes e os resultados coletados experimentalmente habilitam o microinversor flyback a realizar a conversÃo CC-CA de sistemas fotovoltaicos de pequenas potÃncias conectados em cargas lineares monofÃsicas com caracterÃsticas senoidais ou interligados à rede elÃtrica. / The study proposes the design and implementation of a single phase inverter of single stage obtained the derivation of four flybacks converters CC - CC operating in discontinuous mode and associated in parallel modules. This arrangement reduces the amount of magnetic and decreases current efforts in primary transformers and minimizes fluctuations i n photovoltaic module. Among the main properties of micro inverter flyback (MIF) this research, the primary keys work with SPWM switching at high frequency and lagged in one hundred and eighty degrees, while the secondary keys are controlled by PWM's compl ementary. To validate the research, we will develop a prototype that will effect the processing power connecting a photovoltaic panel with 200W power in resistive loads with senoidais characteristics. The proposed system is based on the configuration of th e integrated module drive (MIC), which, in turn, has lower losses in the semi - conductors and magnetic elements of the topology, when compared to complex centralized photovoltaic systems. Among the applications of micro inverter flyback are: supply of remote systems, power electronics devices, small water pumping stations, connections to the grid, among others. Despite the low power, micro inverter has electrical characteristics essential for the insertion in the energy matrix, such as: insulation betwe en the source and the load, types of voltages and currents waves in output with little distortion and frequency of the network. The results obtained in the laboratory show that the micro inverter flyback operating in open loop gets yields close to 90% bein g delivered to the load about 180W. Thus, the ease of construction, the use of fewer components and results collected experimentally enable the micro inverter flyback to perform the CC - CA conversion of photovoltaic systems of small powers connected in sing le - phase linear loads with senoidais characteristics or interconnected to the electric grid

Submitted by Luciana Ferreira (lucgeral@gmail.com) on 2016-09-09T12:27:04Z No. of bitstreams: 2 Tese - Tharsis Souza Silva - 2016.pdf: 3242823 bytes, checksum: 4cdf7de6c7ba7cfe6f4fc07cc9501592 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Luciana Ferreira (lucgeral@gmail.com) on 2016-09-09T12:27:26Z (GMT) No. of bitstreams: 2 Tese - Tharsis Souza Silva - 2016.pdf: 3242823 bytes, checksum: 4cdf7de6c7ba7cfe6f4fc07cc9501592 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2016-09-09T12:27:26Z (GMT). No. of bitstreams: 2 Tese - Tharsis Souza Silva - 2016.pdf: 3242823 bytes, checksum: 4cdf7de6c7ba7cfe6f4fc07cc9501592 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2016-05-13 / Fundação de Amparo à Pesquisa do Estado de Goiás - FAPEG / In this work we _rstly study a relay system X on the Rn that, under certain conditions, it has a one parameter family of 1-periodic orbits that arises in the origin and increase inde_nitely. We study yet another relay system class X_, that it is formed from the initial relay system by aditions of nilpotent parameters that, under certain conditions, it has the same result of the previous, and yet family of periodic orbits that arises in the origin and ends in a loop, or family that bifurcate of a loop and arise inde_nitelly. Furthermore the periodic solutions are explicitely given by Euler polynomials. Finally we study a third order di_erential equation with relay looking for periodic orbits of di_erent degrre of di_erentiability and this is done by the associated vector _eld with jump. / Neste trabalho estudamos primeiramente um campo vetorial descontínuo com chaveamento X atuando no Rn que, sob certas condições, possui uma família a um parâmetro de órbitas 1-periódicas que surge na origem e cresce indenidamente. Estudamos também uma classe de campos vetoriais descontínuos com chaveamento (relay systems) X, que se diferencía do campo inicial pela adição de parâmetros i;j de forma linear Nilpotente que, sob certas condições, possui o mesmo resultado que o caso anterior, e ainda famílias que surgem na origem e termina em um Laço ou mesmo que bifurcam de um laço e crescem indenidamente. Além disso as soluções periódicas são dadas explicitamente através dos polinômios de Euler. Ainda estudamos uma equação diferencial de terceira ordem com chaveamento a m de buscar órbitas periódicas de diferentes graus de diferenciabilidade e esse estudo é feito através do campo vetorial associado com impulso.

Estuda-se o problema de equilíbrio sem força de corpo de uma esfera anisotrópica sob compressão radial uniformemente distribuída sobre o seu contorno no contexto da teoria da elasticidade linear clássica. A solução deste problema prediz o fenômeno inaceitável da auto-intersecção em uma região próxima ao centro da esfera para uma dada faixa de parâmetros materiais. Sob o contexto de uma teoria de minimização do funcional de energia potencial total da elasticidade linear clássica com a restrição de que o determinante do gradiente da função mudança de configuração seja injetivo, este fenômeno é eliminado. Aplicam-se duas formulações do Método dos Elementos Finitos de Galerkin Descontínuo (MEFGD) para obter soluções aproximadas para o problema de equilíbrio da esfera sem restrição. A primeira formulação do MEFGD aproxima diretamente os campos de deslocamento e deformação infinitesimal. A consideração do campo adicional de deformação na formulação do MEFGD aumenta o número de graus de liberdade associados aos nós da malha de elementos finitos e, consequentemente, o custo computacional. Com o objetivo de reduzir o número de graus de liberdade, introduz-se neste trabalho uma formulação alternativa do MEFGD. Nesta formulação, o campo de deformação infinitesimal não é obtido diretamente da inversão do sistema de equações resultante, mas sim por pós-processamento, a partir do campo de deslocamento aproximado. As soluções aproximadas obtidas com ambas as formulações do MEFGD são comparadas com a solução exata do problema sem restrição e com soluções aproximadas obtidas com o Método dos Elementos Finitos de Galerkin Clássico (MEFGC). Ambas as formulações do MEFGD fornecem melhores aproximações para a solução exata do que as aproximações obtidas com o MEFGC. Os erros entre a solução exata e as soluções aproximadas obtidas com a formulação alternativa do MEFGD são um pouco maiores do que os erros correspondentes obtidos com a formulação original do MEFGD. Este aumento nos erros é compensado pelo menor esforço computacional exigido pela formulação alternativa. Este trabalho serve de base para o estudo de problemas com restrição de injetividade utilizando o método de Galerkin descontínuo. / The equilibrium problem without body force of an anisotropic sphere under radial compression that is uniformly distributed on the sphere\'s boundary is investigated in the context of the classical linear elasticity theory. The solution of this problem predicts the unacceptable phenomenon of self-intersection in a vicinity of the center of the sphere for a given range of material parameters. This phenomenon can be eliminated in the context of a theory that minimizes the total potential energy of classical linear elasticity subjected to the restriction that the deformation field be injective. Two formulations of the Finite Element Method using Discontinuous Galerkin (MEFGD) are used to obtain approximate solutions for the unconstrained problem. The first formulation of the MEFGD approximates both the displacement and the strain fields. The consideration of the strain as an additional field in the formulation of the MEFGD increases the number of degrees of freedom associated to the finite elements and, therefore, the computational cost. With the objective of reducing the number of degrees of freedom, an alternative formulation of the MEFGD is introduced in this work. In this formulation, the strain field is not obtained directly from the inversion of the resulting linear system of equations, but from a post-processing calculation using the approximate displacement field. The approximate solutions obtained with both formulations of the MEFGD are compared with the exact solution of the problem without restriction and with approximate solutions obtained with the Finite Element Method using Classical Galerkin (MEFGC). Both formulations of the MEFGD yield better approximations for the exact solution than the approximations obtained with the MEFGC. The errors between the exact solution and the approximate solutions obtained with the alternative formulation of the MEFGD are slightly higher than the corresponding errors obtained with the original formulation of the MEFGD. These errors are compensated by the fact that the alternative formulation requires less computational effort than the computational effort required by the original formulation. This work serves as a basis for the study of problems with the injectivity restriction using the discontinuous Galerkin method.

Pour la conception des cœurs de réacteurs de 4ème génération, une précision accrue est requise pour les calculs des différents paramètres neutroniques. Les ressources mémoire et le temps de calcul étant limités, une solution consiste à utiliser des méthodes de raffinement de maillage afin de résoudre l'équation de transport des neutrons. Le flux neutronique, solution de cette équation, dépend de l'énergie, l'angle et l'espace. Les différentes variables sont discrétisées de manière successive. L'énergie avec une approche multigroupe, considérant les différentes grandeurs constantes sur chaque groupe, l'angle par une méthode de collocation, dite approximation Sn. Après discrétisation énergétique et angulaire, un système d'équations hyperboliques couplées ne dépendant plus que de la variable d'espace doit être résolu. Des éléments finis discontinus sont alors utilisés afin de permettre la mise en place de méthodes de raffinement dite hp. La précision de la solution peut alors être améliorée via un raffinement en espace (h-raffinement), consistant à subdiviser une cellule en sous-cellules, ou en ordre (p-raffinement) en augmentant l'ordre de la base de polynômes utilisée.Dans cette thèse, les propriétés de ces méthodes sont analysées et montrent l'importance de la régularité de la solution dans le choix du type de raffinement. Ainsi deux estimateurs d'erreurs permettant de mener le raffinement ont été utilisés. Le premier, suppose des hypothèses de régularité très fortes (solution analytique) alors que le second utilise seulement le fait que la solution est à variations bornées. La comparaison de ces deux estimateurs est faite sur des benchmarks dont on connaît la solution exacte grâce à des méthodes de solutions manufacturées. On peut ainsi analyser le comportement des estimateurs au regard de la régularité de la solution. Grâce à cette étude, une stratégie de raffinement hp utilisant ces deux estimateurs est proposée et comparée à d'autres méthodes rencontrées dans la littérature. L'ensemble des comparaisons est réalisé tant sur des cas simplifiés où l'on connaît la solution exacte que sur des cas réalistes issus de la physique des réacteurs.Ces méthodes adaptatives permettent de réduire considérablement l'empreinte mémoire et le temps de calcul. Afin d'essayer d'améliorer encore ces deux aspects, on propose d'utiliser des maillages différents par groupe d'énergie. En effet, l'allure spatiale du flux étant très dépendante du domaine énergétique, il n'y a a priori aucune raison d'utiliser la même décomposition spatiale. Une telle approche nous oblige à modifier les estimateurs initiaux afin de prendre en compte le couplage entre les différentes énergies. L'étude de ce couplage est réalisé de manière théorique et des solutions numériques sont proposées puis testées. / The different neutronic parameters have to be calculated with a higher accuracy in order to design the 4th generation reactor cores. As memory storage and computation time are limited, adaptive methods are a solution to solve the neutron transport equation. The neutronic flux, solution of this equation, depends on the energy, angle and space. The different variables are successively discretized. The energy with a multigroup approach, considering the different quantities to be constant on each group, the angle by a collocation method called Sn approximation. Once the energy and angle variable are discretized, a system of spatially-dependent hyperbolic equations has to be solved. Discontinuous finite elements are used to make possible the development of $hp-$refinement methods. Thus, the accuracy of the solution can be improved by spatial refinement (h-refinement), consisting into subdividing a cell into subcells, or by order refinement (p-refinement), by increasing the order of the polynomial basis.In this thesis, the properties of this methods are analyzed showing the importance of the regularity of the solution to choose the type of refinement. Thus, two error estimators are used to lead the refinement process. Whereas the first one requires high regularity hypothesis (analytical solution), the second one supposes only the minimal hypothesis required for the solution to exist. The comparison of both estimators is done on benchmarks where the analytic solution is known by the method of manufactured solutions. Thus, the behaviour of the solution as a regard of the regularity can be studied. It leads to a hp-refinement method using the two estimators. Then, a comparison is done with other existing methods on simplified but also realistic benchmarks coming from nuclear cores.These adaptive methods considerably reduces the computational cost and memory footprint. To further improve these two points, an approach with energy-dependent meshes is proposed. Actually, as the flux behaviour is very different depending on the energy, there is no reason to use the same spatial discretization. Such an approach implies to modify the initial estimators in order to take into account the coupling between groups. This study is done from a theoretical as well as from a numerical point of view.