[pt] No projeto de tubulações que interligam vasos de pressão, reatores, etc., é comum a utilização de curvas para atenuar os esforços que atuam nas paredes dos vasos e reatores. Esses esforços que atuam nas paredes dos vasos e reatores. Esses esforços ocorrem devido aos deslocamentos gerados pela dilatação térmica dos componentes da tubulação. Essas curvas devem, então, possuir grande flexibilidade para a acomodação dos deslocamentos tornando, por isso, menores os esforços nos vasos e reatores. A flexão da curva, provocada pelos deslocamentos, faz surgir, entretanto, tensão normal circuferenciais e longitudinais relativamente grandes quando comparadas à teoria de flexão simples de vigas. Essas tensões são ocasionadas por um fenômeno de ovalização que ocorre em curvas lisas e gomadas. Este trabalho teve como objetivo o estudo da ovalização, das tensões geradas nos tubos curvos compostos por trechos retos (tubos gomados), e da avaliação da confiabilidade dos resultados obtidos para essas tensões. Este estudo foi feito através de uma análise pela teoria de Kitching para tubos gomados; um modelo experimental em forma de U, construído a partir de um tubo de aço de baixo carbono (diâmetro externo de 54mm e espessura de 2.0mm), com uma curva de 176.6mm de raio, analisado pelo método de extensometros elétricos e, finalmente, um modelo analisado pelo Método dos Elementos Finitos. Os resultados obtidos por estes três métodos foram, então, comprados entre si. / [en] Support end reactions that occur in pipelines joining pressure vessels, reactors and others plant equipaments are usually minimized by the introduction of bends in the pipelines. These effects arise due to thermal displacements of the componentes. Therefore, bends are designed to increase the flexibility of the pipelines, making them absorb the thermal displacements in order to decrease the stresses and deflections in the walls of the vessels and reactors. Longitudinal and hoop normal stresses that arise in these bends are relatively high when they are compared to nominal these bends are relatively high when they are compared to nominal stresses, calculated by the ordinary theory of bending of beams. These stresses are induced by a phenomena of flattening, which occurs in smooth and mitred curve bends. The objective of this thesis was to compare results for the stresses induced by the flattenuing effect, determined by three different and independent methods. These three methods were based on an analytical analysis by the kitching theory of flexure of mitred bends, an experimental analysis by the straingage method through an 180 graus mitred bend model made of a low carbon steel tube (64mm external diameter and 2.0 mm thickness) with a bend of 176.6 mm in radius and, finally, a numerical analysis of this tube by the finite element method.

This thesis has identified the failure and damage processes in a particle filled epoxy which is typical of adhesives used industrially. Micromechanical analyses have been carried out to predict the material properties of damaged adhesive, and to investigate the applicability of different failure criteria. The general body of evidence suggests that there is no direct method of predicting the failure load of adhesive joints from the strength and toughness of the adhesive used. Therefore, a favoured approach has been to postulate a failure criterion, and to implement it in the constitutive equation for the adhesive. In contrast, this work has begun from the microstructural modelling of damage, and derived credible failure criteria from this model. The experimental program quantified the adhesive morphology and identified the damage processes that occur in the adhesive prior to failure. Bulk and joint specimens were tested both in-situ in a scanning electron microscope, and on a conventional tensile testing machine. The tests showed that the mechanisms for damage and failure in both joint and bulk form are particle debonding followed by cracking in the matrix. The concept of a representative unit cell of material was used to determine the effects of particle cracking and debonding. In a regular' array of cracked particles, the stiffness remained relatively unchanged in the plane of the cracks, but perpendicular to it, a significant reduction was found. Modelling debonded particles is more complex, because partial contact must be considered in addition to the fully bonded and fully debonded conditions. The unit cell was used to define the elasticity matrix for adhesive containing debonded particles as a function of strain state. The unit cell concept was extended further by including material that obeyed a modified (i.e. hydrostatically sensitive) Von Mises yield criterion. Particle debonding was found to contribute significantly to the hydrostatic sensitivity and to the softening of the adhesive. The unit cell concept was used to implement a strain at a distance failure criteria, using both elastic and plastic material properties. New types of failure criteria also based on the unit cell have been proposed. The criteria relate the strain state in an adhesive joint to the likelihood of shear banding or tensile plastic flow. The regions in a joint that experience one or the other of the mechanisms were identified. Hence the nature and extent of the adhesive failure in joints with varying joint geometry and loading may be predicted.

The overarching objective of this dissertation was to study groundwater resources on global, local, and regional scales. The first objective of this dissertation was to analyze the groundwater nitrate contamination in the Edwards-Trinity and the Southern High-Plains aquifers of Texas. The second was to study groundwater quality in terms of seawater intrusion in the California Coastal Basin, Upper Floridian, and North Atlantic Coastal Plain aquifers. This dissertation also provided a comprehensive overview of the groundwater level in basins at the global scale and further analyzed agricultural activities on groundwater storage in small and large basins. To achieve first objective, Ordinary Least Square (OLS) and Geographically Weighted Regression (GWR) models were used to study the relationship between groundwater nitrate contamination and land use. This dissertation further identified dominant groundwater types using USGS well data and to estimate the extent of seawater intrusion in terms of dominant ions and ocean salinity in the United States coastal aquifers. Finally, groundwater storage anomaly was quantified using Gravity Recovery and Climate Experiment (GRACE) derived variations in total Terrestrial Water Storage (TWS) and the Global Land Data Assimilation System (GLDAS). Land cover data representing a percentage of irrigated lands using groundwater resources was used to study agricultural activities on groundwater storage. Groundwater nitrate contamination was positively associated with cotton production in Southern High-Plains and Edwards-Trinity aquifers. The nitrate concentrations tended to increase as the well-depth decreased in both aquifers. Results showed that the dominant ions in the study area were Na+ and Cl- . The study concluded that Na-Cl and mixed Ca-Mg-Cl were dominant water types in the United States' coastal aquifers. Results also indicated that seawater intrusion is occurring in the US coastal aquifers. Groundwater depletion has increased in southern Asia, western North America, and southwestern Europe due to groundwater withdrawal for agricultural use. However, farming practice is not the main reason for groundwater scarcity in South America, Africa, and Australia.

No description available.

The general problem we consider in this thesis is the following: we have to analyze a stream of data (records, packets, events ...) by successively applying to each piece of data a set of ``rules'. Rules are best viewed as lightweight parallel processes synchronizing on each arrival of a new piece of data. In many applications, such as signature-based intrusion detection, only a few rules are concerned with each new piece of data. But all other rules have to be executed anyway just to conclude that they can ignore it. Our goal is to make it possible to avoid this useless work completely. To do so, we perform a static analysis of the code of each rule and we build a decision tree that we apply to each piece of data before executing the rule. The decision tree tells us whether executing the rule or not will change anything to the global analysis results. The decision trees are built at compile time, but their evaluation at each cycle (i.e., for each piece of data) entails an overhead. Thus we organize the set of all computed decision trees in a way that makes their evaluation as fast as possible. The two main original contributions of this thesis are the following. Firstly, we propose a method to organize the set of decision trees and the set of active rules in such a way that deciding which rules to execute can be made optimally in O(r_u), where r_u is the number of useful rules. This time complexity is thus independent of the actual (total) number of active rules. This method is based on the use of a global decision tree that integrates all individual decision trees built from the code of the rules. Secondly, as such a global tree may quickly become much too large if usual data structures are used, we introduce a novel kind of data structure called sequential tree that allows us to keep global decision trees much smaller in many situations where the individual trees share few common conditions. (When many conditions are shared by individual trees the global tree remains small.) To assess our contribution, we first modify the implementation of ASAX, a generic system for data stream analysis based on the rule paradigm presented above. Then we compare the efficiency of the optimized system with respect to its original implementation, using the MIT Lincoln Laboratory Evaluation Dataset and a classical set of intrusion detection rules. Impressive speed-ups are obtained. Finally, our optimized implementation has been used by Nicolas Vanderavero, in his PhD thesis, for the design of stateful honeytanks (i.e., low-interaction honeypots). It makes it possible to simulate tens of thousands hosts on a single computer, with a high level of realism.

No description available.

No description available.

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Les Loricariinae appartiennent à la famille des poissons-chats néotropicaux cuirassés Loricariidae, la famille de poissons-chats la plus riche en espèce au monde, et se caractérisent par un pédoncule caudal long et aplati et par l’absence de nageoire adipeuse. Préalablement aux études évolutives réalisées, une phylogénie exhaustive et robuste a été établie sur la base de données mitochondriales et nucléaires. Cette phylogénie a ensuite été utilisée dans des analyses multivariées et multi-tableaux afin de révéler les principales tendances évolutives de la sous-famille. La phylogénie obtenue indique que la tribu Harttiini forme un groupe paraphylétique et est restreinte à trois genres, et que dans la tribu Loricariini, deux sous-tribus soeurs se distinguent, les Farlowellina et les Loricariina, chacune présentant des patterns évolutifs complexes. Plusieurs nouveaux taxa ont aussi été mis en évidence et décrits. En utilisant la phylogénie comme outil exploratoire, nous avons démontré : (1) avec l’analyse de co-inertie que les caractères diagnostiques fournis pour définir les différents genres étaient sous dépendance phylogénétique ; (2) avec l’analyse de co-inertie multiple que les forces évolutives sous-jacentes dirigeant leur diversification incluaient des composantes intraphénotypiques (morphologie et génétique) et extraphénotypique (écologie et distribution) ; (3) avec l’analyse RLQ que des évènements de co-dispersion entre espèces codistribuées avaient eu lieu et étaient responsables de la distribution actuelle des espèces ; et (4) avec l’analyse de patterns multi-échelles que la co-évolution des traits liés aux caractéristiques de la bouche était liée à des fonctions reproductrices responsables d’une évolution tertiaire de cet organe. / The Loricariinae belong to the Neotropical mailed catfish family Loricariidae, the mostspeciose catfish family in the world, and are united by a long and flattened caudal peduncle and the absence of an adipose fin. Despite numerous works conducted on this group, no phylogeny is presently available. Prior to conduct evolutionary studies, an exhaustive and robust phylogeny was reconstructed using mitochondrial and nuclear data. Then, this phylogeny was used in multivariate and multi-table analyses to reveal the main evolutionary trends of the subfamily. The resulting phylogeny indicated that the Harttiini tribe, as classically defined, formed a paraphyletic assemblage and was restricted to three genera, and within the Loricariini tribe, two sister subtribes were distinguished, Farlowellina and Loricariina, both displaying complex evolutionary patterns. In addition several new taxa were highlighted and described. Subsequently using this phylogeny as exploratory tool, we demonstrated: (1) using co-inertia analysis that the diagnostic features provided to define the different genera were phylogenetically dependent; (2) using multiple co-inertia analysis that the underlying evolutionary forces shaping their diversification included intraphenotypic (morphology and genetics) and extraphenotypic (ecology and distribution) components; (3) using the RLQ analysis that co-dispersion events occurred between co-distributed species responsible for the current fish distribution; and (4) using the multi-scale pattern analysis that the co-evolution in traits related to the mouth characteristics was linked to reproductive functions responsible for a tertiary evolution of this organ.

Delineating and diagnosing depression and anxiety in the presence of a medical condition, such as diabetes, is complicated by the presence of overlapping symptoms that, therefore, are etiologically ambiguous. These overlapping symptoms include feelings of fatigue, concentration difficulties, restlessness, changes in appetite, irritability and autonomic arousal. The difficulty in understanding these overlapping symptoms has been proposed to lead to an underdiagnosis of depression and anxiety disorders among adults with diabetes, which is problematic given that such disorders are associated with poorer health outcomes. The goal of the current study is to test whether or not these overlapping symptoms are affected by the presence of diabetes by comparing structural models of these symptoms in adults with diabetes versus those free of major medical conditions. Participants include 226 adults with diabetes and 379 adults free of diabetes who completed a series of questionnaires assessing symptoms of depression, anxiety and health status. In addition, for adults with diabetes, the most recent hemoglobin A1c lab result was collected from patient medical records. Results indicate that overlapping symptoms were strongly related to mood for adults with and without diabetes. In conclusion, it is recommended that when these overlapping symptoms are present in adults with diabetes, depression and anxiety should be considered as possible contributors to their presence.