The purpose of this thesis is to analyze how a Breitbart News Network text represented Muslim and Syrian refugees in the weeks leading up to the 2016 United States presidential election. Using a methodological framework based on Norman Fairclough's Critical Discourse Analysis, the analysis seeks to reveal the linguistic semiotic choices made by the text producer in the representation of Muslim and Syrian refugees and to explain how these semiotic choices were used to achieve their effect. By revealing these linguistic semiotic choices, the beliefs, attitudes, and ideologies that are present in the text─ but not necessarily easily seen─ can be brought out into the open. Once out in the open, they can be critically questioned, contested,  and they can be examined to find out how their beliefs, attitudes, and ideologies can be understood to have destructive consequences on the audience and the groups of people they represent.

Les enjeux économiques et sécuritaires majeurs portant sur l'évaluation de la pérennité des systèmes de précontrainte dans les ouvrages d'art conduisent au développement de nouvelles méthodes d'auscultation. Ces travaux représentent la contribution du LGCgE au programme national de recherche ANR ACTENA sur l'auscultation des câbles tendus non-accessibles. Les objectifs de ce travail sont la détermination du tracé des câbles de précontrainte et la détection des zones de défauts d'injection par thermographie infrarouge active. Le premier chapitre rappelle le contexte de cette étude, ainsi que les travaux antérieurs. L'introduction d'un mode de chauffage par induction conduit à une indétermination sur la présence éventuelle de défaut. Afin de lever cet artefact, le deuxième chapitre introduit une méthode originale de traitement fréquentiel des thermogrammes bruts basée sur une approche systémique dans le domaine fréquentiel. Une modélisation par éléments finis des spécimens expérimentaux permet d'étudier la diffusion et d'élaborer les paramètres expérimentaux. Le troisième chapitre expose les résultats expérimentaux obtenus sur trois spécimens de poutres. La procédure de traitement fréquentiel est appliquée aux séries de thermogrammes bruts obtenus. Les résultats sont discutés et mettent en avant l'intérêt des images de phase. Une approche quantitative est appliquée aux résultats de thermographie infrarouge dans le quatrième chapitre. La méthode d'inversion présentée permet la détermination de la diffusivité thermique du béton d'enrobage ainsi que son épaisseur. / The economical and security aspects of the durability of the prestressed concrete construction works lead to develop new non destructive testing methods. This work was developed within the framework of the ANR ACTENA French research project. The main arms are to contribute to the localization of prestressed tendon ducts or rebars and also to the detection of the poor filling defects. The context of this study and previous works are described in the chapter I. The introduction of an inductor heating provides a non uniformity of the heating. In order to avoid any misinterpretation of thermograms, a frequential method based on transfer functions is proposed in chapter II. A 3D numerical model of the experimental specimens is performed by finite element method. Chapter III shows experimental results for three concrete beam specimens. The frequential procedure is applied to the raw thermograms. The results are discussed and highlighted the interest of phase images. A quantitative approach is applied to infrared thermographical results in the chapter IV. Finally, an inverse method is proposed and allowed to the determination of thermal diffusivity and the thickness of the cover concrete.

Cette thèse s'intéresse à l'évaluation de la qualité de l'adhésion dans des assemblages structuraux métal/adhésif/composite carbone-époxy, assemblages utilisés dans l'industrie automobile et aéronautique. La qualité du collage est évaluée par méthodes non destructives grâce à des ondes guidées ultrasonores de type Lamb et SH. Un nouveau modèle numérique par éléments finis a été développé au cours de ce travail : il s'agit d'un modèle rhéologique aux interphases qui permet de tracer les courbes de dispersion des modes guidés dans les échantillons étudiés. Il permet aussi de prédire l'évolution de l'amplitude de ces modes et la répartition de l'énergie dans la structure suivant le niveau d'adhésion ou la présence d'un défaut de type kissing bond. Des échantillons à niveau d'adhésion connus et maîtrisés sont réalisés avec l'aide de physico-chimistes, où l'époxy est partiellement ou totalement réticulé, et où l'interface substrat/adhésif a subi différents traitements chimiques. En particulier sont étudiés des échantillons dont la dégradation du collage dans la structure n'est pas symétrique. Les résultats expérimentaux sont confrontés à ceux issus du modèle numérique. Ces résultats montrent qu'il est possible de caractériser deux niveaux d'adhésion proches et aussi de déterminer sur quelle interface (métal/adhésif ou composite/adhésif) le collage a été dégradé. L'utilisation d'un banc de mesure à transducteurs électromagnétiques acoustiques (EMAT) a permis l'étude de l'amplitude des modes SH. Des variations importantes de l'amplitude de certains modes ont été constatées lors du passage de l'onde d'une zone de bonne adhésion à une zone où le collage est dégradé. / This thesis focuses on the evaluation of the quality of adhesion in metal/adhesive/carbon-epox: structural assemblies used in the automotive and aerospace industry. The quality of the bonding i assessed by non-destructive methods using ultrasonic guided waves of the Lamb and SH types. A nev finite element numerical model was developed during this work: it is an interphase rheological mode that allows the dispersion curves of the guided modes to be plotted in the samples studied. It als( makes it possible to predict the evolution of the amplitude of these modes and the distribution o energy in the structure according to the level of adhesion or the presence of a kissing bond type defect. Samples with known and controlled adhesion levels are made with the help of physico chemists, 'where the epoxy is partially or totally crosslinked, and where the substrate/adhesivl interface has undergone different chemical treatments. In particular, samples are studied whos degradation of the bonding in the structure is not symmetrical. The experimental results are compare( with those from the numerical model. These results show that it is possible to characterize two clos levels of adhesion and also to determine on which interface (metal/adhesive or composite/adhesive the bonding has been degraded. The use of a measurement bench with electromagnetic acousti( transducers (EMAT) allowed thè study of the amplitude of the SH modes. Significant variations in th( amplitude of some modes have been observed when the wave passes from an area of good adhesioi to an area where the bonding is degraded.

Cette thèse vise à développer et à évaluer une méthode de décomposition de domaine avec recouvrement dans le cadre de la modélisation du contrôle non destructif (CND) par courants de Foucault (CF). L'objectif d'une telle approche consiste à éviter le remaillage systématique de l'intégralité du domaine d'étude lors du déplacement de l'un de ses éléments constitutifs(par exemple, déplacement de la sonde CF au dessus de la pièce contrôlée). Plus précisément, il s'agit de concevoir une méthode de décomposition de domaine avec recouvrement qui s'appuie sur la théorie apportée par la méthode des éléments finis avec joints. En plus de s'affranchir de la contrainte d'une interface d'échange invariante avec le mouvement, la technique décrite dans ce travail réalise des transferts d'information réciproques entre les domaines. Cette étude présente les résultats théoriques ainsi que numériques liés à la simulation magnétodynamique. Par ailleurs, l'intérêt d'une telle méthode est illustré par des applications sur des configurations bidimensionnelles de CND par CF. / This thesis aims at studying and developing a domain decomposition method with overlapping subdomains for the modeling in eddy current (EC) non-Destructive testing (NDT). The idea behind such an approach is the possibility to avoid the systematic remeshes of the whole studying domain when some of its components are modified (for example the displacement of the coil above the conductor). More precisely, this work aims at designing a domain decomposition method with overlapping based on the theory of the mortar finite element method. In addition to remove the constraint owing to an coupling interface which is invariant with the displacement, the technique described, in this work, realizes reciprocal transfers of information between subdomains. This study presents the theoretical and numerical results attached to the magnetodynamic simulation. Moreover, the interest of such a method is illustrated by applications in some 2D modeling cases of EC NDT.

Pour répondre à des impératifs croissants de fiabilité et de sûreté, les procédés mis en œuvre dans le cadre du contrôle non destructif sont en constante évolution. Au vu de la complexité des techniques utilisées, la simulation prend une part importante dans leur développement. Nous présentons des travaux ayant abouti à un outil rapide de simulation du champ émis par un traducteur ultrasonore plan quelconque dans des configurations complexes de contrôle non destructif impliquant des géométries maillées sans arêtes saillantes, des matériaux isotropes ou anistropes, homogènes ou hétérogènes et des trajectoires d'ondes pouvant comporter des rebonds et des transmissions. Les fronts d'ondes ultrasonores sont approximés à l'aide d'interpolateurs polynomiaux locaux à des pinceaux de rayons ultrasonores. Ceux-ci sont obtenus au moyen d'un algorithme de recherche de surface par lancer de pinceaux et subdivisions successives. Ils permettent le calcul des grandeurs utiles à la constitution de la réponse impulsionnelle en chaque point d'un échantillonnage du traducteur respectant le critère de Shannon. De cette façon, nous pouvons calculer une réponse impulsionnelle qui, convoluée au signal d'excitation du traducteur, donne le champ ultrasonore. Les performances des simulations ont été accrues par l'exploitation du parallélisme de tâches et des instructions SIMD dans les parties les plus coûteuses du calcul. Enfin, un outil de calcul progressif continu a été développé pour permettre une visualisation interactive d'images de champ. Il exploite une méthode de reconstruction d'images et ordonnance les calculs de champ de manière à accélérer la convergence des images produites / In order to fulfill increasing reliability and safety requirements, non destructive testing techniques are constantly evolving and so does their complexity. Consequently, simulation is an essential part of their design. We developed a tool for the simulation of the ultrasonic field radiated by any planar probes into non destructive testing configurations involving meshed geometries without prominent edges, isotropic and anisotropic, homogeneous and heterogeneous materials, and wave trajectories that can include reflections and transmissions. We approximate the ultrasonic wavefronts by using polynomial interpolators that are local to ultrasonic ray pencils. They are obtained using a surface research algorithm based on pencil tracing and successive subdivisions. Their interpolators enable the computation of the necessary quantities for the impulse response computation on each point of a sampling of the transducer surface that fulfills the Shannon criterion. By doing so, we can compute a global impulse response which, when convoluted with the excitation signal of the transducer, results in the ultrasonic field. The usage of task parallelism and of SIMD instructions on the most computationally expensive steps yields an important performance boost. Finally, we developed a tool for progressive visualization of field images. It benefits from an image reconstruction technique and schedules field computations in order to accelerate convergence towards the final image

Les travaux proposés portent sur la caractérisation par rayons X des matériaux, en vue du tri de certains d'entre eux dans un flux de déchets. L'objectif final du tri est la valorisation de certains matériaux par le biais des circuits de recyclage. Deux méthodes d'analyse utilisant les rayons X sont décrites : la méthode par transmission bi-énergie et la fluorescence X. Ces deux méthodes sont abordées sous un angle théorique et expérimental. Le mémoire se compose de trois parties. La première partie présente le contexte général de l'étude, les différents scénarios de tri envisagés, les différentes méthodes existantes ainsi que les machines de tri industrielles déjà commercialisées. La seconde partie est consacrée à l'analyse par transmission X bi-énergie. Dans cette partie nous présentons tout d'abord le principe de la décomposition d'un matériau dans une base de deux autres matériaux. Ensuite nous utilisons un code de simulation analytique développé au laboratoire afin d'une part d'optimiser la géométrie du détecteur mais également de prévoir les performances de cette méthode en fonction des scénarios et des matériaux recherchés. Nous avons introduit un critère permettant de quantifier la performance d'un scénario donné. Cette méthode a permis de montrer que plusieurs scénarios de tri et en particulier celui du tri des retardateurs de flamme dans un flux de plastiques sont envisageables par cette méthode. Un prototype de machine industrielle a été construit par la société Bertin Technologies et installé dans les locaux de la société Pellenc ST pour réaliser des tests en conditions réelles de tri. Les résultats de simulation ont été comparés aux essais en conditions réelles. Cette partie se conclue avec une introduction aux méthodes multi-énergie. La troisième et dernière partie s'intéresse à la fluorescence X. L'objectif est de mettre au point une méthode permettant d'identifier la nature du matériau en tenant compte des contraintes industrielles (temps d'acquisition faible), c'est-à-dire lorsque le niveau de signal contenu dans le spectre de fluorescence est faible et bruité. Un banc expérimental a été mis au point au laboratoire et a permis de définir la limite de détection qui dépend du matériau analysé et du temps d'acquisition pour réaliser la mesure. Le second point développé dans cette partie est la mise au point d'une technique de simulation de fluorescence combinant le calcul analytique et le calcul par méthode Monte Carlo. L'avantage de cette technique est d'avoir une méthode de simulation plus rapide par rapport à un calcul qui serait réalisé entièrement par Monte Carlo. En conclusion, nous discutons de la complémentarité des informations issues de ces différentes techniques ainsi que leurs avantages et inconvénients. / The proposed work focuses on the characterization of materials by X-ray to sort some of them in a waste stream. The main aim is the valorization of certain of these materials through recycling loops. Two methods of analysis using X-rays are described : dual energy X-ray transmission and X-ray fluorescence. These two methods are discussed from a theoretical and experimental point of view. The memory consists of four parts. The first part presents the general context of the study, the different sorting scenarios considered, the existing methods and industrial sorting systems already in application. The second part is devoted to analysis by dual energy X-ray transmission. In this section we first present the principle of decomposition of a material in a basis of two other materials. Then we use the VXI simulation tool to optimize the geometry of the detector and to predict the performances of the method according to the scenarios and refined materials. This method showed that several sorting scenarios and particularly sorting of flame retardants in plastic flow are possible by this method. An industrial prototype was built by Bertin Technologies and installed in the premises of Pellenc ST to realize sorting tests under actual conditions. Then the results of simulation have been compared to these tests. This section concludes with an introduction of multi-energy methods. The third part deals with X-ray fluorescence. The aim is to develop a method to identify the nature of the material even if the signal contained in its fluorescence spectrum is low and noisy. A experimental bench was developed in the laboratory to define the limit of detection wich depends on the material analyzed and the acquisition time of the measurement. The second point developed in this part is the development of a simulation technique that combines fluorescence analytical calculation and Monte Carlo calculation. The advantage of this technique is to have a faster simulation method compared to a calculation that would be fully realized by Monte Carlo. In conclusion, we discuss the complementarity of information extracted from these different techniques and their respective advantages and drawbacks.

Este trabalho apresenta um método para o reconhecimento e a detecção automática dos diferentes valores ou graus de deformação plástica em Ensaios Não Destrutivos empregando o Ruído Magnético de Barkhausen. O método é baseado no uso de uma Rede Neural Probabilística que permite o diagnóstico automático dos diferentes valores de deformação plástica, conteúdo de carbono, estas medidas são procedentes das medições das amostras de placas de aço AISI 1006, 1050 e 1070, esta base de dados foi feita pelo grupo de pesquisadores do Laboratório de Dinâmica e Instrumentação LADIN da Escola Politécnica da USP, departamento da Mecânica. Os excelentes resultados da rede neural probabilística de detectar automaticamente os valores de deformação mostram a efetividade do desempenho da rede neural probabilística que tem um desempenho superior aos métodos não destrutivos tradicionais e que realmente esta nova tecnologia é uma excelente solução para o diagnóstico. / This work presents a method for automatic detection and recognition of different levels or degrees of plastic deformation in Non-Destructive Testing using the Magnetic Barkhausen Noise. The method is based on using a Probabilistic Neural Network that allows the automatic diagnosis of the different values of plastic deformation and carbon content. The measurements corresponds to samples of steel plates AISI 1006, 1050 and 1070, this database was made by the group of researchers from the Laboratory of Dynamics and Instrumentation LADIN the Polytechnic School of USP, Department of Mechanical Engineering. The results show the effectiveness of the probabilistic neural network to automatically detect plastic deformation levels as well as carbon content level. This method has a superior performance in comparison to traditional nondestructive methods.

O IPEN/CNEN-SP fabrica combustíveis para uso no seu reator nuclear de pesquisas, o IEA-R1. Para qualificar os seus combustíveis tem que comprovar o bom desempenho destes sob irradiação. Como o Brasil não possui reator nuclear de pesquisa com altos fluxos de nêutrons ou células quentes apropriadas para a realização de exames pós-irradiação de combustíveis nucleares, o IPEN/CNEN-SP conduziu um programa de qualificação operacional de elementos combustíveis empregando compostos de urânio já internacionalmente testados sob irradiação e qualificados para uso em reatores de pesquisas, obtendo experiência nas etapas de desenvolvimento tecnológico de fabricação de placas combustíveis, irradiação e ensaios não destrutivos pós-irradiação. Foram fabricados e irradiados diretamente no núcleo do IEA-R1, com sucesso, elementos combustíveis contendo dispersões com baixas frações volumétricas de combustível. Entretanto, existem planos no IPEN/CNEN-SP para aumentar a densidade de urânio dos seus combustíveis. O objetivo deste trabalho de tese consistiu no estudo e proposição de aplicação de um conjunto de métodos não destrutivos para a qualificação dos combustíveis a dispersão de U3O8-Al e U3Si2-Al com alta densidade de urânio fabricados no IPEN/CNEN-SP. Para tanto, foram considerados os recursos de irradiação e a aplicação, na piscina do reator IEA-R1, de métodos não destrutivos disponíveis na Instituição. A proposta consiste em especificar, fabricar e irradiar miniplacas combustíveis, nas densidades máximas já qualificadas internacionalmente e efetuar acompanhamento das condições gerais das mesmas, ao longo do período de irradiação, por meio de métodos não destrutivos na piscina do reator. Além dos métodos de inspeção visual e de sipping já empregados na Instituição, foi concluída a infraestrutura para realização de ensaios dimensionais sub-aquáticos para a avaliação do inchamento das miniplacas combustíveis irradiadas. As análises dos resultados darão subsídios para avaliar e decidir a continuidade ou não das irradiações das miniplacas, até que sejam alcançadas as queimas estipuladas para os testes de irradiação no IEA-R1. / IPEN/CNEN-SP manufactures fuels to be used in its nuclear research reactor the IEA-R1. To qualify those fuels, it is necessary to check if they have a good performance under irradiation. As Brazil doesnt have nuclear research reactors with high neutron fluxes, or suitable hot cells for carrying out post-irradiation examination of nuclear fuels, IPEN/CNEN-SP has conducted a fuel qualification program based on the use of uranium compounds, internationally tested and qualified to be used in research reactors, and has gotten experience in the technological development stages for the manufacturing of fuel plates, irradiation and non-destructive post-irradiation testing. Fuel elements containing low volume fractions of fuel in the dispersion were manufactured and irradiated successfully directly in the core of the IEA-R1. However, there are plans to increase the uranium density of these fuels. The objective of this thesis work was to study and to propose a set of non-destructive methods to qualify the dispersions fuels U3O8-Al e U3Si2-Al with high uranium density produced at IPEN/CNEN-SP. For that, the irradiation resources in the IEA-R1, and the application of non-destructive methods in the reactor pool available in the Institution were considered. The proposal is to specify, manufacture and irradiate fuel miniplates in IEA-R1 at the maximum densities, qualified internationally, and to monitor their general conditions during the period of irradiation, using non-destructive methods in the reactor pool. In addition to the non-destructives visual inspection and sipping methods, already used at the Institution, the infrastructure for dimensional sub-aquatic testing to evaluate the swelling of irradiated fuel miniplates was completed. The analyses of the results will provide means to assess and decide whether or not to continue with the irradiation of miniplates, until the desired burnup for the irradiation tests at IEA-R1 are reached.

Neste trabalho são apresentadas algumas extensões de modelos de sobrevivência com fração de cura, assumindo o contexto em que as observações estão agrupadas. Dois efeitos aleatórios são incorporados para cada grupo: um para explicar o efeito no tempo de sobrevida das observações suscetíveis e outro para explicar a probabilidade de cura. Apresenta-se uma abordagem clássica através dos estimadores REML e uma abordagem bayesiana através do uso de processos de Dirichlet. Discute-se alguns estudos de simulação em que avalia-se o desempenho dos estimadores propostos, além de comparar as duas abordagens. Finalmente, ilustram-se os resultados com dados reais. / In this work some extensions in survival models with cure fraction are presented, assuming the context in which the observations are grouped into clusters. Two random effects are incorporated for each group: one to explain the effect on survival time of susceptible observations and another to explain the probability of cure. A classical approach through the REML estimators is presented as well as a bayesian approach through Dirichlet Process. Besides comparing both approaches, some simulation studies which evaluates the performance of the proposed estimators are discussed. Finally, the results are illustrated with a real database.

Neste trabalho se estuda a aplicação de um método de Ensaio Não Destrutivo, baseado no Ruído Magnético de Barkhausen (RMB), na medição de falhas superficiais em aços. O RMB é gerado por abruptas mudanças na magnetização de materiais quando submetidos a campos magnéticos variáveis. Essas mudanças são afetadas pela microestrutura e a presença e distribuição de tensões elásticas (compressão e tração). Inicialmente apresenta-se um estudo de medições de tensões produzidas por flexão, e perfis de dureza, via RMB. Analisa-se a influência de parâmetros de medição e analise dos sinais, na qualidade do diagnostico. Analisou-se a sensibilidade dos sinais de RMB na medição de tensão a parâmetros tais como direção de laminação da chapa, intensidade e frequência de campo magnético de excitação. Nos estudos da aplicação do RMB em medições de dureza, se usaram amostras de ensaio Jominy, com variação contínua de dureza. Utilizou-se dois tipos de materiais e foram feitas medidas de dureza HRC, RMB e análise metalográfica. O comportamento de diferentes parâmetros dos sinais de RMB se correlacionou com as medições de Dureza e as micrografias obtidas. As amplitudes do RMB mostraram ser inversamente proporcionais aos níveis de dureza. Também se apresenta uma nova tecnologia para inspeção de superfícies, baseado no conceito do Barkhausen Contínuo. São apresentadas análises de parâmetros de medição e de configuração de sonda, na inspeção sob três situações: presença de defeitos volumétricos, deformação plástica, e tensões mecânicas aplicadas. Estudou-se a influência de parâmetros como, amplitude do campo aplicado, velocidade de varredura, posicionamento e características do sensor, no nível de detecção do dano. Métodos de processamento de sinais específicos foram desenvolvidos. Os resultados mostraram, que para cada um dos casos estudados, é possível detectar a posição e nível do dano produzido. Essa nova tecnologia aumenta o espectro de soluções de Ensaios Não Destrutivos para problemas não contemplados pelos métodos existentes. / This work studies the application of a non-destructive testing method, based on the Magnetic Barkhausen Noise (MBN). MBN is generated by abrupt changes in magnetization of ferromagnetic materials subjected to variable magnetic fields. These changes are affected by the presence and distribution of elastic stresses (compression and traction) in the material and by its microstructure. At first, the present study describes MBN measurements of stresses produced by bending and MBN measurements of hardness profiles, in steel samples. The influence of both, measurement parameters and signal analysis, in the quality of fault diagnosis is also analysed. In this context, regarding stress measurements, the MBN signal sensitivity to parameters like surface lamination direction, intensity and frequency of magnetic excitation field is studied. On the other hand, application of MBN to hardness measurements used Jominy essay samples presenting continuous hardness variation. In this case, samples of two different materials were employed and hardness measurements were obtained by HRC, MBN as well as by metallographic analysis techniques. Behaviour of different MBN signal parameters were correlated with obtained hardness measurements and micrography. MBN amplitudes were shown to be inversely proportional to hardness levels. Finally, the work describes a new surface inspection technology, based on the concept of Continuous Barkhausen. In this context, the study analyses probe configuration and measurement parameters, concerning inspection of surface fault from three different origins: volumetric, plastic deformation, applied mechanical stress. The influence of parameters like applied field amplitude, probe velocity, sensor position and characteristics, in detecting damage level was also evaluated. In particular, specific signal processing methods were developed. Results showed that, for each studied case, it is possible to detect damage position and level. This new technology increases the range of NDT essay solutions for problems not yet contemplated by existing methods.