Lorsque des champs électriques sont appliqués à des mélanges eau-huile, les petites gouttelettes d'eau sont attirées entre elles et se regroupent en gouttes plus grosses. Ce processus d’électrocoalescence rend plus efficace la séparation huile-eau par sédimentation.Des données expérimentales sur l’électrocoalescence de très petites gouttelettes sont nécessaires pour améliorer la compréhension de la dynamique de l'interface eau-huile et pour valider les modèles numériques. La configuration simple étudiée dans ce travail de thèse concerne une petite paire de gouttelettes tombant dans une cuve d'huile modèle et soumise à un champ électrique aligné avec l’axe de symétrie des gouttes et la gravité.La première partie du travail a consisté à générer de façon contrôlée d’une paire de très petites gouttelettes (dans la gamme de diamètres 20-200 microns) alignée avec le champ électrique. La génération de goutte à la demande, par méthode éléctrohydrodynamique (EHD) a été améliorée pour un meilleur contrôle du diamètre et de la charge électrique des gouttelettes injectées à partir d'une aiguille métallique unique. Ceci a été obtenu en appliquant à un ménisque d'eau pendant à l’extrémité de l’aiguille des impulsions électriques de forme optimisée.La caractérisation électrique et hydrodynamique des paires de gouttelettes et leur coalescence sont alors principalement déduites de l'analyse des vitesses de chute, avec et sans application d’un champ électrique à courant continu. Des données complètes de positions des gouttelettes et de leur vitesse en fonction du temps sont déduites de prises de vues vidéo. Une attention particulière a été accordée aux visualisations de très petites gouttelettes tombant à petites vitesses, associant des angles multiples de prise de vue, de forts zooms et des vidéos à grande vitesse.La modélisation des différents termes d'interactions hydrodynamiques et électrostatiques entre les gouttelettes permet de déduire des vitesses enregistrées leur masse charge électrique respectives. Quand se produit une coalescence des deux gouttelettes, un enregistrement de la vitesse de la gouttelette résultante, avec et sans tension électrique appliquée, permet de contrôler la conservation de la masse et de la charge électrique, et la validation du procédé.Un premier ensemble de données est constitué d'environ 70 cas différents, avec différentes paire des gouttelettes (dans une plage de diamètre limitée de façon à ce que les vitesses de chute soient comprises entre 0,1 et 0,3 mm/s) et en faisant varier la tension appliquée à courant continu ou alternatif. L'analyse des résultats et des incertitudes expérimentales et un exemple de comparaison possible avec des simulations numériques utilisant le logiciel Comsol Multiphysics ™, permettent d'effectuer des recommandations pour les travaux futurs.Ce travail a été financé par le projet “Fundamental understanding of electrocoalescence in heavy crude oils”; coordonné par SINTEF Energy Research. Le projet a été soutenu par The Research Council of Norway, dans le cadre du contrat n °: 206976 / E30, et par les partenaires industriels suivants: Wärtsilä Oil & Gas Systems AS, Petrobras et Statoil ASA. / When electric fields are applied in oil-water mixtures small water droplets are attracted to others and merge in larger drops. This electrocoalescence process makes more efficient the oil-water separation by sedimentation.Experimental data on the electrocoalescence of very small droplets will be useful to improve the understanding of the dynamics of water-oil interface and to validate numerical models. The simple configuration studied consists in a small droplet pair falling in stagnant model oil, under electric field aligned with the symmetry axis of the droplet pair and the direction of gravity.First part of the work consisted in the well-controlled generation of very small droplet pair (range 20-200 microns) aligned with electric field. Droplet-on-Demand generation by EHD method was improved for a better control of the diameter and electric charge of droplets injected from a single metallic needle. This was obtained by applying to a pendant water meniscus optimized multistage high voltage electric pulses.Electrical and hydrodynamic characterization of the droplet pairs and their coalescence are then mainly deduced from the analysis of falling velocities, with and without applied DC electric field. A complete data set of droplet position and velocity is deduced from video. A special attention was paid to the visualizations of very small droplet and small falling velocities, involving multiple angle of view, strong zooming and high speed video.Modelling the different terms of hydrodynamic and electrostatic interactions between droplets allows deducing from the recorded velocities their respective mass and electric charge. When coalescence occurs, a record of the resulting single droplet velocity, with and without applied voltage, allows controlling the mass and charge conservations and validating the method.A first data set was constituted of about 70 different cases, with varying droplets pair (with a limited diameter range to remain with falling velocities between 0.1 and 0.3 mm/s) and varying applied DC or AC voltage. Analyses of the results and experimental uncertainties, and example of possible comparison with numerical simulations using Comsol Multiphysics™ software, allow performing some recommendations for future work.This work was funded by the project “Fundamental understanding of electrocoalescence in heavy crude oils”; co-ordinated by SINTEF Energy Research. The project was supported by The Research Council of Norway, under the contract no: 206976/E30, and by the following industrial partners: Wärtsilä Oil & Gas Systems AS, Petrobras and Statoil ASA.

Dans un contexte de changement climatique, l'intégration des énergies renouvelables aux systèmes électriques est un enjeu majeur des décennies à venir. Les énergies liées au climat (photovoltaïque, éolien et hydro-électricité) peuvent contribuer à une réduction des émissions de gaz à effet de serre. Cependant, elles sont fortement intermittentes et la production électrique associée peine à répondre à la demande.Cette étude vise à évaluer la faisabilité météorologique du développement d'un système de production électrique basé sur les sources d'énergie liées au climat (CRE - Climate-Related Energy). Nous considérons uniquement leurs variations spatiotemporelles et supposons un équilibre entre production et demande moyennes. Nous avons développé CRE-Mix, une chaîne de modèles permettant de convertir les variables météorologiques en chroniques énergétiques. Cet outil permet l'estimation des fluctuations spatiotemporelles de production et de demande énergétiques résultant de la variabilité hydro-climatique. Pour une sélection de régions en Europe, nous évaluons la facilité d'intégration des CRE en fonctions de leur cohérence temporelle avec la demande. Pour chaque source d'énergie et de multiples mix énergétiques nous estimons successivement (i) le taux de pénétration moyen (PE), qui quantifie la proportion de demande satisfaite sur une longue période et (ii) les caractéristiques des périodes de faible pénétration pour lesquelles le taux journalier de demande satisfaite reste bas pendant plusieurs jours consécutifs. Les résultats montrent que les systèmes basés sur une seule source ont du mal à répondre à la demande et souffrent de longues périodes de faible PE, en raison de leur variabilité temporelle. Cependant, une combinaison d'énergies, l'utilisation de systèmes de stockage ou l'échange d'énergie entre régions, permettent d'augmenter fortement la fiabilité des CRE (PE proche de 100% et rares/courtes périodes de faible pénétration). Cette étude, basée sur 30 ans, a été étendue à l'ensemble de XXème siècle afin d'évaluer les fluctuations basse fréquence des CRE résultant de la variabilité interne du climat. De longues chroniques régionales de production et de demande ont été générées grâce au développement d'une méthode de descente d'échelle statistique basée sur les analogues atmosphériques (SCAMP). Cet outil génère des scénarios météorologiques multivariés physiquement cohérents. Les résultats montrent que les variations basse fréquence des CRE sont influencées par les grandes oscillations océano-climatiques. De plus, on montre que les variations multi-décennales de l'hydro-électricité sont particulièrement importantes avec notamment une différence en PE supérieure à 15% d'une décade à l'autre et des périodes de faible pénétration aux caractéristiques très irrégulières.Enfin, nous évaluons la pertinence de systèmes électriques basés sur les CRE en climat futur. SCAMP permet de produire des scénarios régionaux de variables météorologiques à partir des modèles climatiques issus des simulations CMPI5. Pour les précipitations, les tendances simulées par SCAMP sont en désaccord avec de nombreuses études. L'application de SCAMP en "modèle parfait" semble indiquer que le lien entre les situations atmosphériques de grande échelle et les précipitations totales, mais également convectives et stratiformes, change en climat futur. / In the context of climate change, the integration of renewables in electric power systems is one of the main challenges of the coming decades. Climate-Related-Energy sources (CRE - solar, wind and hydro power) can contribute to reduce the greenhouse gas emissions. However, they exhibit large spatio-temporal fluctuations and the associated intermittent electricity generation often leads to an incomplete supply-demand balance. This study aims to evaluate the meteorological feasibility of developing an electric power system that would only rely on CRE sources. We focus on the multi-scale spatio-temporal fluctuations of these renewables by assuming a balance between mean electricity production and mean energy load. We develop and use CRE-mix, a suite of models able to convert meteorological conditions into CRE time series. It gives an assessment the spatio-temporal fluctuations of power production and energy demand, resulting from the multi-scale hydro-climatic variability. For a set of European regions, we assess the ease of integration of CRE sources, regarding their temporal consistency with energy demand. For each CRE source and multiple CRE mixes, we consider in turn (i) the mean penetration rate (PE), which quantifies the proportion of satisfied demand over a long period and (ii) the characteristics of low penetration periods, defined as sequences of days for which the penetration rate is lower than a given threshold. This study proves that single CRE sources have difficulty to meet the energy demand and suffer from long low penetration periods, due to their multi-scale temporal variations. However, using some integrating factors (multi-sources, storage systems, inter-regions electric power transmission), efficiently improves the reliability of CRE-based power systems with PE rates close to 100% and rare low penetration periods.These analyses, based on a 30-yr period, are extended to the entire 20th century in order to assess the low frequency fluctuations of CRE sources resulting from the internal variability of climate. Long regional series of production and demand, were generated thanks to the development of a statistical downscaling method based on atmospheric analogues (SCAMP). It simulates physically-consistent multivariate series of meteorological parameters. The results demonstrate that these fluctuations are related to some large scale oceano-climatic oscillations. Moreover, the multi-decennial variations of hydro power are particularly large: changes in PE rates exceeding 15% from one decade to the other and uneven energy droughts characteristics.Finally, we evaluate the relevance of the CRE sources under future climate conditions. SCAMP is used to produce downscaled projections of meteorological drivers of CRE sources for the 21st century from a selection of CMIP5 climate models. The resulting scenarios for precipitation are not consistent with other studies focusing of the future modifications of this variable in Europe. The application of SCAMP in a perfect-model approach seems to indicate that the large-scale-meteorology/local-precipitation relationship is changing in the course of the 21st century, for all total, convective and stratiform precipitation.

Cette thèse s’intéresse à la fabrication de pièces composites par le procédé Resin Transfert Molding (ou RTM), appliquée à des tubes de protection thermiques. Plus particulièrement, cette thèse vise à démontrer la faisabilité d’utiliser ce procédé pour la fabrication cette pièce complexe. La phase d’imprégnation de préformes sèches est plus particulièrement étudiée. Après mise en oeuvre, cette pièce peut présenter des défauts tels que de la porosité ou des déplacements de plis constituant la préforme. L’objectif de cette thèse est donc de comprendre l’origine de ces défauts et de minimiser voire de d’empêcher leur apparition. Pour cela, une démarche expérimentale a été mise en place. Celle ci comprend la réalisation d’un pilote de laboratoire permettant d’appliquer différentes conditions d’imprégnation aux préformes considérées. La perméabilité des renforts considérés a aussi été évaluée à différentes échelles grâce à l’utilisation de moyen dédiés à l’échelle macroscopique (banc de perméabilité planaire et transverse), et grâce à l’utilisation d’un code de calcul se basant sur des images de tomographie synchrotron à l’échelle microscopique. Enfin, une analyse de la qualité des prototypes réalisés a été menée en suivant des procédures mises en place lors de ce projet et les résultats analysés et mis en relation avec les conditions de mise en oeuvre. Cette approche expérimentale est couplée aux simulations numériques de la phase d’imprégnation que nous avons aussi mise en oeuvre au cours de cette thèse. Par l’utilisation combinée de la simulation numérique et des essais expérimentaux, nous avons défini des critères estimant le risque d’apparition des défauts. Ces critères ont montré leur efficacité sur les solutions innovantes que nous avons proposées puisque répondant aux exigences du cahier des charges industriel. / This work concerns the manufacturing of composite parts using the Resin Transfer Molding (RTM) process. A major goal of this study is to test the feasibility of using this process to manufacture a thick tubular part with a complex shape. This study concerns the different stages of the process with an important focus on the injection step of dry preforms. The goal of this thesis is to understand the generation of manufacturing defect (mainly porosity and preform deformation) that possibly takes place during the injection step to avoid them. An experimental procedure is proposed. An experimental setup was developed to study the influence of the different process parameters on the quality of the composite parts. The determination of the longitudinal and through the thickness permeabilities was conducted experimentally on sheared and un-sheared samples. An alternative technique to estimate the permeability is presented based on simulation software using X-ray tomography images at the microscale. At last, a quality control procedure was developed and applied to the tubes manufactured within this project. This experimental work was compared to numerical simulations of the injection stage. Using both numerical simulations and experiments, criteria on process and material parameters to predict the quality of the tailored parts are presented. Those criteria are successfully compared to experimental data and were applied to design innovative injection solutions that meet industrial specifications.

Le potentiel de production d'hydroélectricité à l'aide de micro centrales hydroélectriques est sous-exploité en Europe. L'exploitation de petites chutes d'eau, à l'aide des vis d'Archimède, est un moyen de production d'énergie renouvelable récent et en voie de développement. Cette thèse étudie les performances hydrauliques de ce type de turbine. Une première partie du travail a porté sur la mise en place d'un dispositif expérimental, à échelle réduite, permettant de tester pour des conditions hydrauliques et des paramètres géométriques variés les performances hydrauliques d'une vis d'Archimède. Un modèle théorique semi-analytique permettant de déterminer les performances de la vis a été développé puis validé à l'aide des résultats expérimentaux. Les écoulements complexes au sein de cette turbine sont étudiés par le biais d'une modélisation numérique 3D instationnaire, validée expérimentalement. La compréhension de la structure de l'écoulement a alors permis de développer un nouveau modèle théorique permettant de déterminer avec plus de précision le débit de fuite qui est à l'origine d'une perte significative de rendement. Ces résultats permettent d'envisager l'implémentation d'un modèle de dimensionnement industriel. / The potential for hydropower generation using micro-hydro plants is still under exploited in Europe. The Archimede Screw Generators are a growing technology convenient for low-head hydraulic sites. This thesis studies the hydraulic performance of this turbine. The first part of thiswork presents an experimental device using a laboratory screw scale. This device allows to test screw performance for various hydraulic conditions and geometrical parameters. A theoretical model predicting the screw performance has then been developed and validated with experimental results.The complex flows occurring within the screw are studied thanks to 3D and unsteady numerical simulations. The numerical results are validated experimentally. The insights provided on the flow structure permit to develop a new leakage model. These leakages are a major source of efficiency loss. Eventually, all these results will allow the implementation of an industrial dimensioning model.

Os processos de soldagem por atrito apresentam diversas vantagens na união de materiais metálicos. O Friction Hydro Pillar Processing (FHPP) é um processo recente utilizado para fabricação e reparo de estruturas, que surge como uma alternativa à soldagem convencional. As interfaces das soldas FHPP são geralmente as partes mais suscetíveis à falha, devido à concentração de defeitos e microestrutura heterogênea. Por ser uma técnica recente e pouco difundida é importante avaliar as juntas soldadas por esse processo. Diante desse contexto, este trabalho teve como objetivo analisar os resultados macro e microestruturais de um aço C-Mn submetido ao processo FHPP, utilizando-se diferentes parâmetros, e relacioná-los com o perfil de microdureza. Além disso, avaliar a tenacidade ao impacto nas regiões de interface da solda e no material base. Para isso, foram realizados estudos metalográficos e perfis de microdureza Vickers em diferentes regiões das soldas. Complementarmente foram ensaiadas as interfaces das soldas que obtiveram os melhores resultados, através do ensaio Charpy, para avaliação de tenacidade ao impacto. Durante a soldagem foram utilizadas diferentes forças axiais e de recalque, obtendo como resultados variações, principalmente, nas interfaces soldadas. As alterações visualizadas nas interfaces, geraram uma tenacidade ao impacto menor do que a do material base. Esse fato pode ser atribuído, principalmente, porque o material base apresentou uma microestrutura predominantemente ferrítica, enquanto as interfaces soldadas apresentaram microestruturas heterogêneas, com presença de ferrita acicular, ferrita de Widmanstätten, martensita revenida e bainita. / Friction welding processes offer several advantages to join metallic materials. Friction Hydro Pillar Processing (FHPP) is a recent process used to manufacture and repair of underwater structure, emerging as an alternative for the conventional welding. FHPP welding interface are generally most susceptible to failure due to its heterogeneous microstructure and defects concentration. As FHPP is a quite recent little known technic, evaluate joints welded by this process is important. In this context, this study objective is to analyze macro and microstructural results of a C-Mn steel welded under different FHPP process parameters, and relate this results to a microhardness profile. Besides that, evaluate welding interface and base material impact toughness. To do that, metallographic analysis and microhardness profiles were done in different regions of the welded joint. In addition welded interfaces with the best results were Charpy tested for impact toughness evaluation. Different axial and forging forces were used, achieving macro and microstructural variations mainly in welded interface as a result. These variations caused interface to have lower impact toughness than base material. It can be explained by the base material mainly ferritic structure, whilst welded interface showed heterogeneous microstructures with acicular ferrite, Widmantätten ferrite, tempered martensite and bainite.

O processo de soldagem por fricção com pino consumível possui como campo de aplicação a área de reparo e manutenção de estruturas. Por ser uma técnica desenvolvida recentemente, o conhecimento relativo à sua aplicação ainda não está completamente fundamentado. Dessa forma é necessário que se entenda o comportamento das juntas soldadas para determinar a viabilidade de sua utilização. Atualmente uma das ferramentas existentes para isso é o emprego dos conceitos de mecânica da fratura, que possibilitam a determinação de valores quantitativos de tenacidade. A partir disso objetiva-se identificar qual é a resposta frente à tenacidade à fratura que diferentes materiais desempenham quando soldados por fricção com pino consumível. No presente trabalho foi utilizada cpmo parâmetro de análise a medição de tenacidade à fratura pelo método de CTOD (Crack Tip Opening Displacement) em juntas soldadas com diferentes materiais de pino consumível e metal base. Essas medições concentraram-se nos materiais originais do pino, do metal de base e na região adjacente à interface lateral das juntas. Os materiais estudados foram o aço BS EN10025, os inoxidáveis austeníticos AISI 316L e ASTM F138, e o duplex UNS S31803. Adicionalmente foram realizadas análises fractográfica, macro e microestrutural, bem como ensaios de microdureza visando o entendimento do comportamento das juntas soldadas. Os resultados de CTOD indicaram que as microestruturas originadas com a soldagem, bem como a presença de inclusões de MnS, influenciam significativamente a tenacidade. / The Friction Hydro Pillar Processing has as a field of application in the area of repair and maintenance of structures. However as it is a technique developed recently, knowledge on its application is not fully established. Thus, it is necessary to understand the behavior of the joints in order to determine the feasibility of use. Nowadays one of the existing tools for this is the use of the concepts of fracture mechanics that allow the determination of quantitative values of toughness. From that this work aims to analyse fracture toughness in different materials welded by Friction Hydro Pillar Process. The analysis of fracture toughness was performed by the method CTOD (Crack Tip Opening Displacement) in samples welded with different materials of stud and base material. These measurements were concentrated in the original materials of and the interface zone of the welds. The materials investigated were BS EN10025, austenitic stainless steels AISI 316L and ASTM F138, the duplex stainless steel UNS S31803. Additionally were performed macro and microstructural characterization and microhardness tests aimed to understanding the behavior of joints. The CTOD results indicated that the microstructures as well as the presence of MnS inclusions significantly influence the toughness.

Os processos de união por soldagem entre peças e componentes metálicos são indispensáveis para nossa realidade tecnológica. O processo de soldagem por fricção com pino consumível conhecido como FHPP (Friction Hydro Pillar Processing) é um processo desenvolvido recentemente que ocorre no estado sólido através da fricção entre dois materiais e, com isso, reúne algumas vantagens sobre os processos convencionais de soldagem por fusão. Esta técnica apresenta grande potencial de aplicação na indústria de óleo e gás visando reparos de trincas e descontinuidades em equipamentos e estruturas de aços em geral. Apesar do grande potencial desta técnica, o estudo sobre a sua aplicação para produzir juntas com materiais dissimilares ainda é escasso se comparado com outras técnicas de soldagem. Desta forma, este trabalho tem como objetivo o desenvolvimento e a análise de juntas soldadas utilizando pinos com adição de níquel em chapas de aço com manganês, visando determinar a combinação entre material de pino e parâmetros de soldagem que forneça a junta com melhores características mecânicas e metalúrgicas. Os parâmetros de soldagem que variaram foram: velocidade de rotação e a força axial. Os materiais utilizados nos conjuntos soldados foram chapas de aço ao manganês, classificadas comercialmente como ASTM A516 grau 70, e barras de aço baixo carbono com adição de 3, 6 e 9% de níquel. A análise realizada sobre as juntas abrangeu a caracterização da microestrutura e das propriedades de dureza das regiões específicas da solda além dos ensaios mecânicos de dobramento, tração com micro-amostras e de tenacidade à fratura através do parâmetro CTOD. Todas as soldas tiveram bom preenchimento, porém alguns materiais de pino não proporcionaram boa ductilidade durante o dobramento. As falhas ocorridas durante o dobramento juntamente com os estudos da tenacidade à fratura na zona de ligação mostraram que o acúmulo de inclusões nesta região representa degradação das propriedades mecânicas localmente. Os ensaios de tração com micro-amostras indicam aumento da resistência mecânica na região do material depositado do pino e os valores de CTOD no material termomecânicamente afetado são significativamente maiores do que na zona de ligação, porém não alcançam os valores obtidos nos materiais como não processados. / The processes of union by welding between parts and metallic components are indispensable for our technological reality. The Friction Hydro Pillar Processing (FHPP) process is a newly developed process that occurs in the solid state through friction between two materials and thus has some advantages over conventional fusion welding processes. This technique presents great potential of application in the oil and gas industry aiming repairs of cracks and discontinuities in equipment and steel structures in general. Despite the great potential of this technique, the study of its application to produce joints with dissimilar materials is still scarce compared to other welding techniques. The aim of this work is the development and analysis of welded joints using rod with nickel addition in steel sheets with manganese, in order to determine the combination of rod material and welding parameters that provides the joint with better mechanical and metallurgical characteristics. The welding parameters that varied were: speed of rotation and axial force. The materials used in the welded assemblies were manganese steel plates, commercially classified as ASTM A516 grade 70, and low carbon steel with addition of 3, 6 and 9% nickel. The analysis of the joints included the characterization of the microstructure and the hardness properties of the specific regions of the weld in addition to the bend test, micro-tensile test and fracture toughness through the parameter CTOD. All welds were well filled, however, some rod materials did not provide good ductility during bend test. The failures during folding together with the fracture toughness studies in the bonding zone showed that the accumulation of inclusions in this region represents degradation of the mechanical properties locally. The micro-tensile test indicate increased mechanical strength in the region of the deposited material of the rod and the values of CTOD in the thermomechanically affected material are significantly higher than in the bonding zone, but do not reach the values obtained in the unprocessed material.

Este estudo tem por objetivo avaliar a possibilidade de confeccionar juntas soldadas pelo processo de soldagem por fricção com pino consumível (Friction Hydro Pillar Processing), em aço classificado como grau R4 de acordo com a norma IAC W22. As juntas soldadas foram produzidas com três diferentes forças axiais: 30 kN, 45 kN e 60 kN, e após a soldagem, submetidos a três diferentes tempos de tratamento térmico de revenimento: uma hora, duas horas e quatro horas, na temperatura de 650 C. A microestrutura predominantemente martensítica resultante destes processos, foi caracterizada através de microscopia óptica, e de um mapeamento de microdureza Vickers. As soldas tiveram suas zonas afetadas pelo processo mensuradas através do software de análise de imagens ImageJ, sendo observadas maiores zonas para menores forças axiais empregadas. Caracterizou-se ainda, o nível de tensões residuais oriundos do processo de soldagem através do método de difração de nêutrons, para uma solda produzida com força axial de 45 kN, em duas diferentes condições, no estado de como soldado e com revenimento de quatro horas após a soldagem. Com os resultados obtidos neste trabalho pode-se concluir que as juntas soldadas com a força de 60 kN apresentaram a melhor combinação de propriedades avaliadas, que são, um menor tamanho de zonas afetadas pelo processo, e manteve um bom nível de dureza mesmo após o revenimento. Todas as juntas soldadas apresentaram uma ótima união metalúrgica, e os diferentes tempos de tratamento térmico empregados promoveram alterações microestruturais e nas propriedades do material já esperadas para o tratamento térmico de revenimento. / The aim of this study is to evaluate the possibility of making joints welded by Friction Hydro Pillar Processing, in steel grade R4 according to IAC W22. The repairs were produced with three different axial forces: 30 kN, 45 kN and 60 kN, and after welding, subjected to three different tempering heat treatment times: one hour, two hours and four hours, at a temperature of 650 C. The predominantly martensitic microstructure resulting from these processes was characterized by optical microscopy and a Vickers microhardness mapping. The welds had their zones affected by the process measured through the ImageJ image analysis software, with larger zones being observed for smaller axial forces employed. It was also characterized the level of residual stresses from the welding process, through the neutron diffraction method, for a weld produced with axial force of 45 kN, in two different conditions, as welded and annealing for four hours after welding. With the results obtained in this work it can be concluded that the welded joints with the force of 60 kN presented the best combination of evaluated properties, which are, a smaller size of zones affected by the process, and maintained a good level of hardness even after the annealing. All the welded joints presented an excellent metallurgical bonding, and the different times of thermal treatment employed promoted microstructural changes and in the properties of the material already expected for the heat treatment of annealing.

Les anciennes peintures sur bois représentent une part importante du patrimoine culturel européen. Depuis longtemps, il est connu que les conditions de stockage peuvent entraîner des dégradations et endommagements aux panneaux. Une bonne compréhension de l'état actuel des peintures sur bois ainsi que la prédiction de leur comportement dans diverses conditions climatiques est essentielle pour mieux les conserver et les restaurer. De telles analyses peuvent être faites par le biais de la modélisation numérique. Cependant, la modélisation numérique est efficace que si les paramètres des matériaux sont équivalents à ceux de la réalité. Dans ce travail, les modifications mécaniques, physiques et chimiques qui se produisent au cours du vieillissement naturel ont été étudiées pour le bois et les couches de peintures. Les possibilités d'accélérer le vieillissement par des traitements thermo-hydrique ont été appliquées avec succès au bois. Une loi de vieillissement du bois a été trouvée pour la résistance radiale et la luminosité de la couleur. Malheureusement, une tel loi n'a pas été trouvé pour les couches de peinture. Ensuite, un modèle numérique tridimensionnel sur le comportement hygroscopique et mécanique des panneaux peints a été développé. La comparaison entre ce modèle et des expériences sont en bon accord. Finalement, une analyse de risques sur une ancienne peinture a montré que des fluctuations d'humidité relative normales ne devraient pas promouvoir de dommages dans le support en bois, mais la fissuration de la couche picturale est possible. / Ancient wood panel paintings represent an important part of the European cultural heritage. Since long time it is known that the storing condition can degrade and damage them. Understanding of the actual state of old panel paintings and the prediction of their behaviour under various climatic conditions is essential to better conserve and restore them. Such analysis could be done by numerical modelling. However, numerical modelling can achieved good predictions only if the materials parameters are adequately determined. In this work, the mechanical, physical and chemical modifications occurring during natural aging have been investigated for wood and paint layers. The possibility to accelerate the aging with thermo-hydro treatments has been successfully applied to wood. A wood aging law has been found for the radial strength and the lightness. Unfortunately, no such law has been found for the paint layers. The hygroscopic and mechanical behaviour of panel paintings has been then simulated in 3D by finite element method model including time dependent behaviour. The comparison between the model and experiments shows good agreement. Finally, risk analysis on an old painting shows that normal climate fluctuation should not promote cracks in the wooden support but the cracking of the ground layer is possible.

Os processos de união por soldagem entre peças e componentes metálicos são indispensáveis para nossa realidade tecnológica. O processo de soldagem por fricção com pino consumível conhecido como FHPP (Friction Hydro Pillar Processing) é um processo desenvolvido recentemente que ocorre no estado sólido através da fricção entre dois materiais e, com isso, reúne algumas vantagens sobre os processos convencionais de soldagem por fusão. Esta técnica apresenta grande potencial de aplicação na indústria de óleo e gás visando reparos de trincas e descontinuidades em equipamentos e estruturas de aços em geral. Apesar do grande potencial desta técnica, o estudo sobre a sua aplicação para produzir juntas com materiais dissimilares ainda é escasso se comparado com outras técnicas de soldagem. Desta forma, este trabalho tem como objetivo o desenvolvimento e a análise de juntas soldadas utilizando pinos com adição de níquel em chapas de aço com manganês, visando determinar a combinação entre material de pino e parâmetros de soldagem que forneça a junta com melhores características mecânicas e metalúrgicas. Os parâmetros de soldagem que variaram foram: velocidade de rotação e a força axial. Os materiais utilizados nos conjuntos soldados foram chapas de aço ao manganês, classificadas comercialmente como ASTM A516 grau 70, e barras de aço baixo carbono com adição de 3, 6 e 9% de níquel. A análise realizada sobre as juntas abrangeu a caracterização da microestrutura e das propriedades de dureza das regiões específicas da solda além dos ensaios mecânicos de dobramento, tração com micro-amostras e de tenacidade à fratura através do parâmetro CTOD. Todas as soldas tiveram bom preenchimento, porém alguns materiais de pino não proporcionaram boa ductilidade durante o dobramento. As falhas ocorridas durante o dobramento juntamente com os estudos da tenacidade à fratura na zona de ligação mostraram que o acúmulo de inclusões nesta região representa degradação das propriedades mecânicas localmente. Os ensaios de tração com micro-amostras indicam aumento da resistência mecânica na região do material depositado do pino e os valores de CTOD no material termomecânicamente afetado são significativamente maiores do que na zona de ligação, porém não alcançam os valores obtidos nos materiais como não processados. / The processes of union by welding between parts and metallic components are indispensable for our technological reality. The Friction Hydro Pillar Processing (FHPP) process is a newly developed process that occurs in the solid state through friction between two materials and thus has some advantages over conventional fusion welding processes. This technique presents great potential of application in the oil and gas industry aiming repairs of cracks and discontinuities in equipment and steel structures in general. Despite the great potential of this technique, the study of its application to produce joints with dissimilar materials is still scarce compared to other welding techniques. The aim of this work is the development and analysis of welded joints using rod with nickel addition in steel sheets with manganese, in order to determine the combination of rod material and welding parameters that provides the joint with better mechanical and metallurgical characteristics. The welding parameters that varied were: speed of rotation and axial force. The materials used in the welded assemblies were manganese steel plates, commercially classified as ASTM A516 grade 70, and low carbon steel with addition of 3, 6 and 9% nickel. The analysis of the joints included the characterization of the microstructure and the hardness properties of the specific regions of the weld in addition to the bend test, micro-tensile test and fracture toughness through the parameter CTOD. All welds were well filled, however, some rod materials did not provide good ductility during bend test. The failures during folding together with the fracture toughness studies in the bonding zone showed that the accumulation of inclusions in this region represents degradation of the mechanical properties locally. The micro-tensile test indicate increased mechanical strength in the region of the deposited material of the rod and the values of CTOD in the thermomechanically affected material are significantly higher than in the bonding zone, but do not reach the values obtained in the unprocessed material.