[pt] Uma das características mais relevantes nos aglomerados de minério de ferro é a sua porosidade, que impacta fortemente no desempenho desses materiais nos processos siderúrgicos. O desempenho é diretamente dependente da existência de uma rede porosa que permite o fluxo de gases pelo interior desses aglomerados sem comprometer sua integridade física. Neste trabalho, amostras de diferentes tipos de aglomerados de minério de ferro foram caracterizadas com o auxílio de técnicas de microtomografia computadorizada de raios X (microCT), processamento digital de imagens e modelagem de rede de poros (PNM). Com isso, a influência da microestrutura desses aglomerados na variação da sua porosidade e permeabilidade foi avaliada. O uso de microCT permitiu uma visualização 3D da estrutura dos aglomerados, permitindo realizar uma análise da estrutura interna das amostras para a discriminação do espaço poroso. O pixel size ideal foi estipulado por meio de diversas capturas com resoluções diferentes. A PNM foi utilizada para realizar a simulação da permeabilidade absoluta das amostras, correlacionando com a porosidade, conectividade dos poros e diâmetro de poros e conexões. Foi realizada uma variação de mais ou menos 5 tons de cinza nos limiares de segmentação para estipular a sensibilidade do impacto desse parâmetro nos resultados da modelagem. Os dois aglomerados apresentaram porosidade parecida, em torno de 20 por cento. Os resultados para piores resoluções apresentaram uma inconsistência, em muitos casos não possuindo sequer permeabilidade. As imagens adquiridas com um tamanho de voxel de 2 micrômetros resultaram em cálculos consistentes de permeabilidade, em torno de 0,4 a 2,4 mD para os briquetes e 0,03 a 1,6 mD para as pelotas, sugerindo que os briquetes são levemente mais permeáveis. A variação do limiar de segmentação dos poros teve forte impacto nos resultados das modelagens, influenciando diretamente no valor do cálculo da permeabilidade absoluta. / [en] One of the most relevant features of iron ore agglomerates is their porosity, which strongly impacts the performance of these materials in steelmaking processes. Performance is directly dependent on the existence of a porous network that allows gas flow through the interior of these agglomerates without compromising their physical integrity. This study characterized samples of different iron ore agglomerates using X-ray microcomputed tomography (microCT), digital image processing, and pore network modeling (PNM). The influence of the microstructure of these agglomerates on the variation of their porosity and permeability was evaluated. MicroCT enabled a 3D visualization of the agglomerate structure, allowing for an analysis of the internal structure of the samples to discriminate the porous space. The ideal pixel size was determined through various captures at different resolutions. PNM was used to simulate the absolute permeability of the samples, correlating it with porosity, pore connectivity, and pore and connection diameter. A variation of more or less 5 gray tones in the segmentation thresholds was performed to determine the sensitivity of this parameter s impact on the modeling results. The two agglomerates had similar porosity of around 20 percent. The results for lower resolutions showed inconsistency, with many cases lacking permeability altogether. Images acquired with a pixel size of 2 micrometers resulted in consistent permeability calculations, ranging from 0.4 to 2.4 mD for briquettes and 0.03 to 1.6 mD for pellets, indicating that briquettes are slightly more permeable. The variation of pore segmentation threshold had a strong impact on the modeling results, directly influencing the value of the absolute permeability calculation.

[pt] O processo de formação de pelotas de minério de ferro consiste na preparação das matérias-primas, formação da pelota crua e endurecimento por meio da queima. O produto final deve ser um material poroso que permita a difusão de gases no forno de redução e que, simultaneamente, resista a compressão, característica relevante durante o transporte e no carregamento do forno. No entanto, durante o tratamento térmico e o transporte podem surgir trincas que comprometem a integridade das pelotas. A discriminação de poros e trincas é, portanto, um importante fator para a análise microestrutural e controle de qualidade do material. A microtomografia de raios-x é uma técnica não destrutiva que gera imagens tridimensionais, o que permite uma visualização completa da pelota. No entanto, a metodologia usual de processamento digital de imagens, baseada em extração de atributos de tamanho e forma, apresenta limitações para discriminar poros de trincas. Redes Neurais Deep Learning são uma alternativa poderosa para classificar tipos de objetos em imagens, utilizando como entrada as intensidades dos pixels e atributos automaticamente determinados pela rede. Após treinar um modelo com os padrões correspondente a cada classe, é possível atribuir cada pixel da imagem a uma das classes presentes, permitindo uma segmentação semântica. Nesta dissertação, otimizou-se uma rede Deep Learning com arquitetura U-Net, usando como conjunto de treinamento poucas camadas 2D da imagem 3D original. Aplicando o modelo à pelota utilizada no treinamento foi possível discriminar poros de trincas de forma adequada. A aplicação do modelo a outras pelotas exigiu a incorporação de camadas destas pelotas ao treinamento e otimização de parâmetros do modelo. Os resultados apresentaram classificação adequada, apesar de apresentar dificuldades de criar um modelo geral para discriminação entre poros e trincas em pelotas de minério de ferro. / [en] The iron ore pellet forming process consists of preparing the raw materials, forming the raw pellet and hardening by firing. The end product must be a porous material which allows gas to diffuse in the blast furnace and at the same time resists compression, which is a relevant feature during transport and loading of the furnace. However, during heat treatment and transport cracks may appear that compromise the integrity of the pellets. The discrimination of pores and cracks is therefore an important factor for microstructural analysis and material quality control. X-ray microtomography is a non-destructive technique that generates three-dimensional images, allowing a full view of the pellet. However, the usual methodology of digital image processing, based on extraction of size and shape attributes, has limitations to discriminate crack from pores. Deep Learning Neural Networks are a powerful alternative to classifying object types in images, using as input the pixel intensities and attributes automatically determined by the network. After training a model with the patterns corresponding to each class, it is possible to assign each pixel of the image to one of the classes present, allowing a semantic segmentation. In this dissertation, a Deep Learning network with U-Net architecture was optimized, using as a training set a few 2D layers of the original 3D image. Applying the model to the pellet used in training it was possible to discriminate cracks pores properly. Application of the model to other pellets required the incorporation of layers of these pellets into the training and optimization of model parameters. The results were adequately classified, despite the difficulty of creating a general model for discrimination between pores and cracks in iron ore pellets.

Ce mémoire présente une méthode mise au point pour caractériser et analyser des fissures de fatigue présentant un fort caractère tridimensionnel dans des matériaux métalliques opaques. L'analyse consiste à déterminer avec précision la position du front de la fissure étudiée et à mesurer des valeurs de facteurs d'intensité des contraintes le long du front par projection sur les séries de Williams du champ de déplacement issu de la corrélation numérique d'images 3D obtenues par tomographie aux rayons X. La corrélation d'images 3D numériques est utilisée afin de mesurer le champ de déplacement en volume lors de la mise sous chargement d'une éprouvette fissurée fatiguée. La corrélation d'images nécessitant un mouchetis, le matériau retenu pour les expériences est la fonte à graphite sphéroïdal car il présente un mouchetis 3D naturel (les nodules de graphites) parfaitement imagé par tomographie aux rayons X. Le cyclage est appliqué à l'aide d'une machine de fatigue in situ permettant d'alterner des phases de propagation de la fissure avec des acquisitions tomographiques sous différentes charges. L'introduction d'un défaut artificiel (une entaille obtenue par usinage laser) permet de maîtriser l'amorçage et la propagation de la fissure in situ. La méthode de corrélation d'images 3D numériques employée dans ces travaux étant basée sur des éléments finis, nous avons cherché à tirer profit de différents outils développés dans le cadre de cette méthode. Les surfaces libres sont spécifiées afin de bien conditionner le maillage et un enrichissement dans l'esprit des X-FEM permet de renseigner la fissure dont la position est repérée grâce à la trace laissée dans le résidu de corrélation entre l'image avant cyclage et la dernière image acquise. Une régularisation mécanique est également introduite dans le calcul sous forme d'un filtre de longueur d'onde choisie. Le champ de déplacement mesuré avec précision est ensuite projeté sur les séries de Williams augmentées des termes correctifs de Leblond et Torlai qui prennent en compte la courbure du front de la fissure. L'annulation du terme super-singulier d'ordre -1 des séries de Williams est utilisée pour détecter la position du front de la fissure. Une procédure itérative a été mise en place afin de concilier l'enrichissement et la courbure du front avec la projection sur les séries de Williams. Une fois la position du front 3D de la fissure déterminée et les valeurs des facteurs d'intensité des contraintes associées calculées, les résultats obtenus sont confrontés à la littérature. / This manuscript describes a methodology used to compute Stress Intensity Factor values along the curved front of a fatigue crack inside a nodular cast iron. An artificial defect is introduced at the surface of a small sample. The initiation and growth of a fatigue crack from this defect during constant amplitude cycling is monitored in situ by laboratory x-ray tomography. The method for processing the 3D images in order to compute SIF values is described in detail. The results obtained show variations of the stress intensity factor values along the crack front.

L’objectif de ce travail est d’étudier les couplages entre les propriétés hygroélastiques des papierset leur microstructure. L’exploitation d’images de papiers modèles acquises par microtomographieà rayons X a permis de caractériser de manière inédite l’évolution des propriétésmicrostructurales de ces matériaux en fonction de leurs conditions d’élaboration ainsique lors d’essais où ils ont été placés sous atmosphère à humidité relative contrôlée.Ces résultats constituent un apport nouveau pour la connaissance de la nature statistique desdescripteurs des propriétés des fibres (dimensions et orientation) et de leurs contacts (surface,degré de liaison), de l’architecture des réseaux fibreux que forment les papiers (nombrede contacts entre fibre) et pour la taille des volumes élémentaires représentatifs des propriétésmicrostructurales et élastiques de ces milieux. Sur la base de ces informations, différentsmodèles, plus oumoins raffinés, ont été élaborés dans le cadre de la théorie de l’homogénéisationdes structures périodiques discrètes, pour décrire les propriétésmécaniques des papiers.Cette approche apporte un éclairage nouveau sur le rôle des liaisons entre fibres sur leur comportement mécanique. / The objective of this work is to study the coupling between the hygroelastic properties ofpapers and their microstructure. The use of images of models acquired by X-ray microtomographypapers allowed the characterization in an unprecedentedmanner of the evolutionofmicrostructural properties of thesematerials according to their production conditions andduring tests where they were placed in atmosphere at controlled relative humidity. These resultsprovide a new contribution to the knowledge of the statistical nature of the descriptorsof fibre properties (size and orientation) and their contacts (surface, bonding degree ratio), ofthe architecture of fibrous networks that papers constitute (number of fibre-to-fibre bonds)as well as of the size of the representative elementary volumes of microstructural and elasticproperties. Based on this information, various models, more or less sophisticated, were developedin the framework of the theory of homogenisation of discrete periodic structures todescribe the mechanical properties of paper. This approach sheds new light on the role offibre-to-fibre bonds on themechanical behaviour of thesematerials.

Dynamic soil water repellency is an important soil phenomenon in the vadose zone as it is now recognised that most soils in the world are likely to express some degree of reduced wettability and/or long term hydrophobicity. Fractional wettability and contact angles are, however, rarely discussed or quantified for natural systems. This is particularly the case in the presence of dynamic contact angles. Soil water repellency remains a persistent impediment and challenge to accurate conceptual and numerical models of flow and storage in the vadose zone. This dissertation addresses the opportunity and pressing need for contributions that develop better quantifiable definitions, descriptions, and understanding of soil water repellency. Using materials collected from post wildfire sites, this work employs water and ethanol to identify, isolate, and quantify contact angle dynamics and fractional wettability effects during infiltration. Varied concentrations of water and ethanol solutions were applied to soils and observed through X-ray microtomography, tension infiltration experiments, and moisture content measurements in the laboratory and field. Several analyses from lab and field investigations showed that applications of ethanol and specifically, water-ethanol aqueous solutions provide unique additional insights into proportions of media that remain non-wettable and how those proportions affect overall hydrologic processes, which are not readily observable through water infiltrations alone. Observations include the wetting up of microporous structures, reduced storage, and changes in unsaturated hydraulic conductivities. Challenges which develop as a consequence of variable fluid properties including changes to operational pore assemblages, slow down of wetting fronts, and non-uniqueness relative to infiltration responses are addressed. Important insights and contributions were developed through this approach and water-ethanol mixtures are valuable tools for developing greater quantification and mechanistic data to better inform our models and understanding of dynamic soil water repellency. / Dissertation / Doctor of Philosophy (PhD) / Quantifying fluid behaviours in soils is important for a host of environmental, social, and economic reasons. Over the last 25+ years, one soil phenomenon has garnered increased attention because it interferes with our ability to carry out this work. Soils that are or become water repellent develop all over the world and where hydrophobic or non-wetting substances can enter soil and remain in pore spaces or as coatings on particles. To assist in the tracking and management of its complex effects on water storage and infiltration, the goals of this work were to develop fundamental insights into the manifestation and effects of this variable soil property on key hydrologic properties and processes. This work tests a new conceptual model for understanding these systems through both field and laboratory work and using a number of different technologies. These include X-ray microtomography (μXCT), tension infiltrometry, and more regularly applied techniques which are sensitive to changes in repellency. The works shows how combining fractional wettability and contact angle dynamics generates a stereoscopic conceptual framework which facilitates increased capacity for quantifying and understanding of soil systems expressing dynamic soil water repellency.