Dissertations / Theses on the topic 'Fusion de traces'

Volatile elements released from the subducting slab play a fundamental role during the formation of arc magmas in the mantle wedge. Advances of melt inclusion studies enlarged the data on volatile abundance in arc magmas, and it is now possible to characterize some volatile contents in arc primary magmas, in particular F and Cl. A recent study of Mt Shasta melt inclusions (LeVoyer et al., 2010) shows that fractionation of F and Cl potentially contains information about arc magma genesis. In order to trace the source of arc magmas, fluorine and chlorine partitioning was investigated. Here, I present new experimental determinations of Cl and F partition coefficients between dry and hydrous silicate melts and mantle minerals: olivine, orthopyroxene, clinopyroxene, plagioclase, garnet and also pargasite and phlogopite. The values were compiled from more than 300 measurements in 24 melting experiments, conducted between 8 and 25 kbars and between 1180 and 1430˚C. The low abundance F, Cl measurements in minerals were done by Cameca IMF 1280 at WHOI using the negative secondary ion mode. The results show that DOpx/meltF ranges from 0.123 to 0.021 and DCpx/meltF ranges from 0.153 to 0.083, while Cl partition coefficient varies from DOpx/meltCl from 0.002 to 0.069 and DCpx/meltCfrom 0.008 to 0.015, as well. Furthermore, DOl/meltF ranges from 0.116 to 0.005 and DOl/meltCl from 0.001 to 0.004; DGrt/meltF ranges from 0.012 to 0.166 and DGrt/meltCl from 0.003 to 0.087 with the increasing water amount and decreasing temperature. I also show that F is compatible in phlogopite DPhl/meltF > 1.2) while DAmp/meltF is incompatible in pargasite DAmp/meltF from 0.36 to 0.63). On the contrary, Cl is more incompatible in phlogopite (DPhl/meltCl > 1.2 on average 0.09 ± 0.02), than in pargasite (DPhl/meltCl from 0.12 to 0.38). This study demonstrates that F and Cl are substituted in specific oxygen site in minerals that lead then to be more sensitive than trace elements to crystal chemistry and water amount variations thus melting conditions. Using those new partition coefficients, I modelled melting of potential sub-arc lithologies with variable quantity aqueous-fluid. This model is able to decipher 1) amount of aqueous-fluid involved in melting, 2) melting induced by fluid or melting of an hydrous mineral-bearing source and 3) melting of either pargasite-bearing lithology or phlogopite-bearing lithology and shows that sources of some primitive melts, for instance from Italy, bear pargasite and phlogopite, while some primitve melts seem to be the results of fluid-induced melts.

Le suivi visuel d’objet est une fonction élémentaire de la vision par ordinateur ayant fait l’objet de nombreux travaux. La dérive au cours du temps est l'un des phénomènes les plus critiques à maîtriser, car elle aboutit à la perte définitive de la cible suivie. Malgré les nombreuses approches proposées dans la littérature pour contrer ce phénomène, aucune ne surpasse une autre en terme de robustesse face aux diverses sources de perturbations visuelles : variation d'illumination, occultation, mouvement brusque de caméra, changement d'aspect. L’objectif de cette thèse est d’exploiter la complémentarité d’un ensemble d'algorithmes de suivi, « trackers », en développant des stratégies de fusion en ligne capables de les combiner génériquement. La chaîne de fusion proposée a consisté à sélectionner les trackers à partir d'indicateurs de bon fonctionnement, à combiner leurs sorties et à les corriger. La prédiction en ligne de dérive a été étudiée comme un élément clé du mécanisme de sélection. Plusieurs méthodes sont proposées pour chacune des étapes de la chaîne, donnant lieu à 46 configurations de fusion possibles. Évaluées sur 3 bases de données, l’étude a mis en évidence plusieurs résultats principaux : une sélection performante améliore considérablement la robustesse de suivi ; une correction de mise à jour est préférable à une réinitialisation ; il est plus avantageux de combiner un petit nombre de trackers complémentaires et de performances homogènes qu'un grand nombre ; la robustesse de fusion d’un petit nombre de trackers est corrélée à la mesure d’incomplétude, ce qui permet de sélectionner la combinaison de trackers adaptée à un contexte applicatif donné<br>Visual object tracking is an elementary function of computer vision that has been the subject of numerous studies. Drift over time is one of the most critical phenomena to master because it leads to the permanent loss of the target being tracked. Despite the numerous approaches proposed in the literature to counter this phenomenon, none outperforms another in terms of robustness to the various sources of visual perturbations: variation of illumination, occlusion, sudden movement of camera, change of aspect. The objective of this thesis is to exploit the complementarity of a set of tracking algorithms by developing on-line fusion strategies capable of combining them generically. The proposed fusion chain consists of selecting the trackers from indicators of good functioning, combining their outputs and correcting them. On-line drift prediction was studied as a key element of the selection mechanism. Several methods are proposed for each step of the chain, giving rise to 46 possible fusion configurations. Evaluated on 3 databases, the study highlighted several key findings: effective selection greatly improves robustness; The correction improves the robustness but is sensitive to bad selection, making updating preferable to reinitialization; It is more advantageous to combine a small number of complementary trackers with homogeneous performances than a large number; The robustness of fusion of a small number of trackers is correlated to the incompleteness measure, which makes it possible to select the appropriate combination of trackers to a given application context

Tracker-level fusion for video tracking combines outputs (state estimations) from multiple trackers, to address the shortcomings of individual trackers. Furthermore, performance evaluation of trackers at run time (online) can determine low performing trackers that can be removed from the fusion. This thesis presents a tracker-level fusion framework that performs online tracking performance evaluation for fusion. We first introduce a method to determine time instants of tracker failure that is divided into two steps. First, we evaluate tracking performance by comparing the distributions of the tracker state and a region around the state. We use Distribution Fields to generate the distributions of both regions and compute a tracking performance score by comparing the distributions using the L1 distance. Then, we model this score as a time series and employ the Auto Regressive Moving Average method to forecast future values of the performance score. A difference between the original and forecast returns the forecast error signal that we use to detect tracking failure. We test the method with different datasets and then demonstrate its flexibility using tracking results and sequences from the Visual Object Tracking (VOT) challenge. The second part presents a tracker-level fusion method that combines the outputs of multiple trackers. The method is divided into three steps. First, we group trackers into clusters based on the spatio-temporal pair-wise relationships of their outputs. Then, we evaluate tracking performance based on reverse-time analysis with an adaptive reference frame and define the cluster with trackers that appear to be successfully following the target as the on-target cluster. Finally, we fuse the outputs of the trackers in the on-target cluster to obtain the final target state. The fusion approach uses standard tracker outputs and can therefore combine various types of trackers. We test the method with several combinations of state-of-the-art trackers, and also compare it with individual trackers and other fusion approaches.

Urban activities involving planning, preparing for and responding to time critical situations often demands sound situational awareness of overall settings. Decision makers, who are tasked to respond effectively to emergencies, must be equipped with information on the details of what is happening, and must stay informed with updates as the event unfolds and remain attentive to the extent of impact the dynamics of the surrounding settings might have. Recent increases in the volumes of geo-spatial data such as satellite imageries, elevation maps, street-level photographs and real-time imageries from remote sensory devices affect the way decision makers make assessments in time-critical situations. When terrain related spatial information are presented accurately, timely, and are augmented with terrain analysis such as viewshed computations, enhanced situational understanding could be formed. Painting such enhanced situational pictures, however, demands efficient techniques to process and present volumes of geo-spatial data. Modern Graphics Processing Units (GPUs) have opened up a wide field of applications far beyond processing millions of polygons. This dissertation presents approaches that harness graphics rendering techniques and GPU programmability to visualize urban terrain with accuracy, viewshed analysis and real-time imageries. The GPU ray tracing and image fusion visualization techniques presented herein have the potential to aid in achieving enhanced urban situational awareness and understanding. Current state of the art polygon based terrain representations often use coarse representations for terrain features of less importance to improve rendering rate. This results in reduced geometrical accuracy for selective terrain features that are considered less critical to the visualization or simulation needs. Alternatively, to render highly accurate urban terrain, considerable computational effort is needed. A compromise between achieving real-time rendering rate and accurate terrain representations would have to be made. Likewise, computational tasks involved in terrain-related calculations such as viewshed analysis are highly computational intensive and are traditionally performed at a non-interactive rate. The first contribution of the research involves using GPU ray tracing, a rendering approach, conventionally not employed in the simulation community in favor of rasterization, to achieve accurate visualization and improved understanding of urban terrain. The efficiency of using GPU ray tracing is demonstrated in two areas, namely, in depicting complex, large scale terrain and in visualizing viewshed terrain effects at interactive rate. Another contribution entails designing a novel approach to create an efficient and real-time mapping system. The solution achieves updating and visualizing terrain textures using 2D geo-referenced imageries for enhanced situational awareness. Fusing myriad of multi-view 2D inputs spatially for a complex 3D urban scene typically involves a large number of computationally demanding tasks such as image registrations, mosaickings and texture mapping. Current state of the art solutions essentially belongs to two groups. Each strives to either provide near real-time situational pictures in 2D or off-line complex 3D reconstructions for subsequent usages. The solution proposed in this research relies on using prior constructed synthetic terrains as backdrops to be updated with real-time geo-referenced images. The solution achieves speed in fusing information in 3D. Mapping geo-referenced images spatially in 3D puts them into context. It aids in conveying spatial relationships among the data. Prototypes to evaluate the effectiveness of the aforementioned techniques are also implemented. The benefits of augmenting situational displays with viewshed analysis and real-time geo-referenced images in relation to enhancing the user's situational awareness are also evaluated. Preliminary results from user evaluation studies demonstrate the usefulness of the techniques in enhancing operators' performances, in relation to situational awareness and understanding.<br>Ph.D.<br>Doctorate<br>Industrial Engineering and Management Systems<br>Engineering and Computer Science<br>Modeling and Simulation

PETRÓLEO BRASILEIRO S. A.<br>A determinação de elementos traços em amostras de rocha por espectrometria de massas, combinada com ablação a laser (LA-ICPMS), tem se tornado uma importante ferramenta nas investigações em petrogênese, metalogênese e na prospecção de minérios. Apesar do surgimento de uma nova geração de lasers com menor duração de pulso (fs) e comprimento de onda (<266 nm), a calibração é ainda um ponto crítico, pois em geral, não há padrões disponíveis com matriz semelhante. Uma das interferências não espectrais mais relevantes em LA-ICPMS é o fracionamento químico, o qual acarreta diferenças entre as composições do alvo (padrão ou amostra) e da massa ablacionada, comprometendo a exatidão e a repetitividade dos resultados. Quando a análise integral é o objetivo principal, o assemelhamento de matriz pode ser alcançado por simples fusão das amostras e padrões em misturas de meta e tetraboratos, de modo similar ao que é feito, rotineiramente, na análise por fluorescência de raios-X (XRFA). Este trabalho relata metodologia para a determinação de elementos traços em amostras de basaltos, provenientes de sítios de exploração de petróleo. Os experimentos foram realizados com um sistema de ablação a laser CETAC LSX-100 em combinação com um espectrômetro ELAN 5000 ICPMS. Uma mistura de argônio-nitrogênio (2 % N2, 98 % Ar) foi utilizada como gás carreador, com o intuito de aumentar a massa ablacionada, e consequentemente, melhorar as intensidades de sinal e reduzir os limites de detecção. Conectou-se uma câmara ciclônica antes da tocha do ICP, com o objetivo de uniformizar o tamanho das partículas de aerossol, e deste modo, melhorar a repetitividade da medição e a robustez do plasma. A título de comparação, análises também foram feitas sem o emprego desta câmara. Padrões de calibração foram preparados a partir de dois materiais de referência certificados, o basalto NIST SRM 688 e a obsidiana NIST SRM 278. Os padrões de rochas pulverizadas foram misturados e homogeneizados com fundente de borato de lítio (CLAISSE, puro, 75 % Li2B4O7, 25 % LiBO2) nas proporções mássicas de rocha: fundente de 1:5, 1:11, 1:23, 1:47, respectivamente, obtendose, deste modo, cada padrão em quatro diferentes concentrações. O padrão interno escolhido foi o In, tendo sido este adicionado antes da fusão, resultando em alvos com a concentração de 100 mg kg(-1) neste elemento. Obtiveram-se os brancos, de modo similar, usando o fundente borato de lítio. A validação metodológica foi feita empregando-se o basalto (GSJ, JB-2) e o granito (NIMG, SARM-1), tendo sido a partir destes, preparados vidros com a proporção mássica (amostra: fundente) de 1:5 e contendo o padrão interno. Todas as fusões foram feitas em um forno de fusão automático (CLAISSE Fluxy). Análises por ICPMS e ICP OES foram feitas após dissolução de fragmentos dos alvos em HNO3, para verificação de possíveis perdas de elementos voláteis. Quarenta elementos foram determinados, a maioria das curvas de calibração apresentou coeficiente de determinação (R2) maiores que 0,995. Limites de detecção variaram de 0,013 mg kg(-1) para Tb, até 0,6 mg kg(-1) para Zn. Estes valores foram melhorados para vários elementos pelo uso da câmara ciclônica (V, Zn, Rb, Y, Nb, Ba, Ce, Sm, Tb, Dy, Ho, Tm). Nas análises do basalto (GSJ, JB-2) e do granito (NIMG, SARM-1), a exatidão foi melhor do que 5 % para V, Sr, Y, Ba, Ce, Nd, Yb e Lu; entre 5-10 % para: Sc, Zn, Rb, Tm, e de 10-20 % para: Co, Zr, La e Tb. A repetitividade foi melhor do que 5% para V, Rb, Sr, La, Nd, Ta; entre 5-10 % para: Sc, Co, Zn, Sr, Y, Zr, Ba, Tb, Yb; e de 10-20 % para: Ce, Tm, Lu e Hf. Foi utilizado ainda o método semiquantitativo TotalQuant II o qual, dentro da incerteza esperada para este tipo de calibração (cerca de 10 % a 20 %), forneceu resultados concordantes com calibração externa quantitat<br>Determination of trace elements in rock samples by laser ablation inductively coupled plasma mass spectrometry (LA-ICPMS) has become an important tool for investigations in petrogenesis, metallogenesis and ore prospecting. However, despite the availability of new generation lasers with shorter pulse duration (fs) and wave length (<266 nm), calibration remains still a critical issue when matrixmatched standards are not available. One of the most serious non- spectral interference in LA-ICPMS is chemical fractionation, which causes compositions of standard and sample and their respective ablated masses to be different, thus compromising accuracy and precision of the results. When bulk analysis is the main objective, matrix matching can be achieved by the simple fusion of samples and standards with meta/tetraborates fluxes, similar to what is routinely used in XRFA. This work reports on our experience with this procedure in the analysis of basaltic rocks, aiming at the trace element characterization of host rocks from oil exploration sites in Brazil. The experiments were performed with a CETAC LSX- 100 laser ablation system (Nd:YAG, 266 nm, operated in the Q-switched and scanning mode) coupled to an ELAN 5000 ICPMS. An argon- nitrogen mixture (2 % N2, 98 % Ar) was used as carrier gas to enhance the mass of material ablated, thus increasing signal intensities and reducing detection limits. A cyclonic spray chamber was arranged just before the ICP-torch as a tentative to uniform aerosol particle size for improving measurement repeatability and robustness of the plasma. For comparison, the system was also operated without a spray chamber. Calibration samples were prepared from two standard reference materials: NIST SRM 688 basalt and NIST SRM 278 obsidian. The powdered rock standards were mixed and homogenized with a lithium borate flux (CLAISSE, pure, 75 % Li2B4O7, 25 % LiBO2) using rock-to-flux mass ratios of 1:5, 1:11, 1:23 and 1:47, respectively, thus obtaining each standard in four different analyte concentrations. Indium was used as an internal standard (IS) and was added before fusion to achieve targets with IS concentration of 100 mg kg(-1). Blanks from the lithium borate flux were produced in a similar way. Method validation was performed with GSJ basalt JB-2 and NIMG granite SARM-1, prepared at a constant sampleto- flux mass ratio of 1L:5 and containing also the IS. All fusions were performed in an automated furnace (CLAISSE Fluxy). Solution nebulization ICP-MS/OES after dissolution of smaller target pieces in HNO3 was used to assess for volatilization losses. Forty elements were determined and good linearity of the calibration curves was obtained and for most of them the correlation coefficient (R2) was higher than 0.995. Detection limits ranged from 0.013 mg kg(-1) for Tb to 0.6 mg kg(-1) for Zn, and improved for several elements by using a cyclonic spray chamber (e.g. V, Zn, Rb, Y, Nb, Ba, Ce, Sm, Tb, Dy, Ho, Tm). In the analysis of GSJ basalt JB-2 and NIMG granite SARM-1, accuracy was better than 5 % for: V, Sr, Y, Ba, Ce, Nd, Yb, and Lu; between 5-10 % for: Sc, Zn, Rb, Tm, and 10-20 % for: Co, Zr, La and Tb. Repeatability was better than 5 % for: V, Rb, Sr, La, Nd, Ta, between 5-10% for: Sc, Co, Zn, Y, Zr, Ba, Tb, Yb, and 10-20 % for: Ce, Tm, Lu and Hf. Additionally, the semiquantitative TotalQuant II calibration was applied, which gave, within the expected uncertainty for this calibration method (10 % to 20 %), concordant results when compared to the quantitative external calibration procedure. Preliminary results on the application of these methods for the determination of trace elements in basaltic host rocks from oil exploration sites will be shown.

L'imagerie fonctionnelle par Tomographie d'Émission de Positons (TEP) multi-traceurs pourrait avoir un rôle essentiel dans le traitement du cancer par radiothérapie externe. Les images TEP aux traceurs 18Fluoro-Déoxy-Glucose (18 FDG), 18F-Fluoro-L Thymidine (18 FLT) et 18Fluoro-Misonidazole (18 FMiso) sont respectivement témoins du métabolisme glucidique, de la prolifération cellulaire et de l'hypoxie (manque d'oxygénation des cellules). L'utilisation conjointe de ces trois traceurs pourrait permettre de définir des sous-volumes donnant lieu à un traitement particulier. À cet effet, il est impératif de mettre à la disposition du corps médical un outil de segmentation et de fusion de ces images. Les images TEP ont pour caractéristique d'être très bruitées et d'avoir une faible résolution spatiale. Ces imperfections ont pour conséquence respective d'induire la présence d'informations incertaines et imprécises dans les images. Notre contribution réside dans la proposition d'une méthode, nommée EVEII pour Evidential Voxel-based Estimation of Imperfect Information, basée sur la théorie des fonctions de croyance, offrant une segmentation fiable et précise dans le contexte d'images imparfaites. Elle réside également dans la proposition d'une méthode de fusion d'images TEP multi-traceur. L'étude d'EVEII sur des images simulées a révélé qu'elle est la mieux adaptée comparée à d'autres méthodes basées sur la théorie des fonctions de croyance, en donnant un taux de bonne reconnaissance des pixels de près de 100 % lorsque le rapport signal-sur-bruit dépasse 2, 5. Sur les fantômes TEP physiques, simulant les caractéristiques des images TEP au 18 FDG, à la 18 FLT et au 18 FMiso, les résultats ont montré que notre méthode estime le mieux les volumes des sphères à segmenter comparé à des méthodes de la littérature proposées à cet effet. Sur les deux fantômes faiblement et fortement bruités respectivement, les biais moyens d'erreur d'estimation des volumes sont seulement de -0,27 et 3,89 mL, témoignant de sa pertinence à visée de segmentation d'images TEP. Enfin, notre méthode a été appliquée à la segmentation d'images TEP multi-traceurs chez trois patients. Les résultats ont montré que notre méthode est adaptée à la fusion d'images TEP multi-traceurs, offrant à cet effet un ensemble d'images paramétriques permettant de différencier les différents tissus biologiques.

Bibliography: pages 200-204.<br>A low dilution fusion technique using a 2:1 flux:sample ratio has been developed for the accurate determination of major, minor and trace elements by x-ray fluorescence spectrometry (XRFS). This method has been used to analyze geological samples of widely varying and unusual composition such as lamproites and kimberlites. The results are shown to be of comparable if not better accuracy than other methods of sample preparation for XRFS. Analytical conditions, including corrections for spectral line interferences are reported for all the elements determined. For major element analysis three methods of calculating appropriate alpha coefficients were investigated, namely (1) multiple regression analysis, (2) Norrish and Hutton method and (3) a computer program, NBSGSC, involving fundamental parameters and the COLA equation. Methods (1) and (2) gave poor results for rock types of widely varying composition. The third method requires that the variation of alpha coefficients with varying weight fraction of the analyte element be taken into account when calculating alpha coefficients if accurate values for major and minor elements are to be obtained on low dilution fusion discs. For trace element analysis the average relative error was less than 5% and there was a decrease in sensitivity by about a factor of 2 compared with XRFS determinations on powder pellets. The elimination of particle size effects in the homogeneous glass fusion discs is a major advantage over the use of powder pellets, especially for the determination of elements such as Ba, Cr and the REEs. A loss on fusion technique has been employed to ensure complete loss of volatiles from the rock samples. Data presented highlight the problems encountered in the determination of the volatile content in geological samples. The oxidation and retention of sulphur in the discs was also investigated.

Le volcanisme des zones de subduction est associé à la déshydratation de la plaque lithosphérique subduite. La présence d'eau permet d'abaisser les conditions nécessaires au déclenchement de la fusion partielle du manteau.<br /><br />Un modèle numérique bidimensionnel en éléments finis a été développé afin de déterminer l'influence des paramètres géométriques de la subduction sur cette fusion partielle. Ce modèle intègre de manière couplée la résolution des équations de la mécanique des fluides, le calcul de la fusion partielle et les transferts d'eau entre les différentes phases minéralogiques du manteau et de la croûte. Du point de vue technique, les équations de conservation de l'énergie sont résolues par une méthode de type Galerkin tandis que les équations de conservation de la masse et du moment utilisent une procédure de pénalisation. La prise en compte des transferts de fluides se fait par le biais d'une méthode de traceurs. Chacun d'entre eux porte une information de composition minéralogique et de teneur en eau. En fonction des conditions de pression et de température, nous pouvons déterminer, à partir d'un diagramme de phase, les teneurs maximales en eau que peut accepter un traceur et donc évaluer la migration du fluide.<br /><br />Les résultats obtenus sont en accord avec les données géochimiques enregistrées en surface dans les laves des arcs<br />volcaniques. Ainsi, une augmentation de l'épaisseur de la croûte chevauchante est associée à une diminution de la production magmatique totale, tandis qu'une augmentation du taux de convergence des plaques est corrélée avec une augmentation du taux de fusion partielle.

Nous avons étudié des magmas ferrifères silicatés magnésiens à la pression du manteau terrestre en utilisant la dynamique moléculaire (First Principles Molecular Dynamics). Les résultats de l’équation d’état que nous avons obtenus à partir de nos simulations ont été utilisés pour créer un modèle chimique et minéralogique pour les zones de très basse vitesse sismique (ULVZ, anomalies régionales dans le manteau proche de la limite noyau-manteau). De plus, nous avons étudié le comportement du Ni, du Co et du Fe dans ces magmas et établi la dépendance du spin en fonction de la concentration, de la pression, de la température et du degré de polymérisation du magma silicaté. Nous avons montré qu’une baisse du spin moyen peut être corrélée au changement de pente (kink) observé précédemment pour les coefficients de partage du Ni et du Co. Nous avons analysé la structure du magma pour toutes les compositions étudiées en fonction de la pression. Nos résultats donnent un nouvel aperçu de la coordination des éléments majeurs et traces dans les magmas silicatés de différents degrés de polymérisation. Nous interprétons l’anomalie de coordination Ni-O en fonction de la pression comme un changement d’état de spin. L’effet de la polymérisation du magma silicaté sur les coefficients de partage du Co, du Ni et du W entre le métal et le magma silicaté a été étudié par expériences multi-enclumes en conditions isobares et isothermes. Nous avons réalisé des simulations FPMD de magmas à des degrés de polymérisation similaires aux expériences afin d’expliquer le caractère de plus en plus lithophile du W lorsque le degré de polymérisation du magma silicaté diminue. Nous proposons une explication structurale pour expliquer l’affinité décroissante apparente du W dans les magmas silicatés dépolymérisés<br>We explore Fe-bearing Mg-silicate melts through the pressure regime of the Earth’s mantle using First Principles Molecular Dynamics (FPMD). The equation of state results we obtained from our simulations are used to create a chemical and mineralogical model for Ultra-Low Velocity Zones (anomalous region on the mantle side of the core-mantle boundary). Furthermore we study the behaviour of Ni, Co, and Fe in these melts, and asses their spin-crossover dependencies on their concentration, pressure, temperature, and the degree of polymerization of the silicate melts. We show that a decrease in the average spin can be correlated with the previously observed kink in the partitioning coefficient of Ni and Co. We investigate the melt structure of all the compositions studied as a function of pressure. Our results provide new insight into the coordination of major and trace elements in silicate melts with different degrees of polymerization. We interpret the anomalous Ni-O coordination trend with pressure as the result of the spin state change. The effect of silicate melt polymerization on the partitioning of Co, Ni, and W between a metal and silicate melt, is investigated at isobaric and isothermic conditions using multi-anvil experiments. We have performed FPMD simulations of melts with similar degrees of polymerization as the experiments in order to explain the increasing lithophile character of W with the decrease in polymerization of the silicate melt. We propose a structural explanation for tungsten’s apparent increased affinity for depolymerized silicate melts

La notion de temps de parcours est une information simple à intégrer par les usagers des transports et a le potentiel, via des systèmes d’assistance, de réduire la congestion aussi bien de manière temporelle que spatiale. Cette thèse porte sur la problématique de l’estimation du temps de parcours en ville, où la circulation se caractérise par un débit variable et un flux significatif de/vers des voies transversales non équipées de capteurs de trafic, flux qui agit comme perturbation de l’équation de conservation des véhicules. La bibliographie présente des approches déterministes et stochastiques, utilisant, selon les situations expérimentales, soit des boucles inductives, soit des véhicules traceurs, soit les deux. Nous proposons une méthode stochastique du type filtrage de Kalman sans parfum. Son modèle dynamique est basé sur la méthode analytique classique des stocks qui considère le décalage temporel entre les cumuls de véhicules entrants et sortants dans chaque tronçon du réseau. La formulation de ce modèle n’est pas explicite, ce qui justifie l’utilisation d’un filtre sans parfum. Les observations utilisées sont à la fois les boucles magnétiques à la sortie des tronçons et les véhicules traceurs après map-matching sur le réseau. L’algorithme est applicable en temps réel, sans connaissance a priori des affectations aux carrefours, et sous conditions éventuelles de perturbation. En sortie de filtre, on dispose des variances d’estimation, ce qui mesure la confiance dans le temps de parcours estimé et permet de rejeter des mesures aberrantes. La validation a été montrée sur la base de simulations sur un réseau simple à trois intersections, avec des flux variables et des perturbations, et pour quelques pourcents de véhicules traceurs.

La notion de temps de parcours est une information simple à intégrer par les usagers des transports et a le potentiel, via des systèmes d'assistance, de réduire la congestion aussi bien de manière temporelle que spatiale. Cette thèse porte sur la problématique de l'estimation du temps de parcours en ville, où la circulation se caractérise par un débit variable et un flux significatif de/vers des voies transversales non équipées de capteurs de trafic, flux qui agit comme perturbation de l'équation de conservation des véhicules. La bibliographie présente des approches déterministes et stochastiques, utilisant, selon les situations expérimentales, soit des boucles inductives, soit des véhicules traceurs, soit les deux. Nous proposons une méthode stochastique du type filtrage de Kalman sans parfum. Son modèle dynamique est basé sur la méthode analytique classique des stocks qui considère le décalage temporel entre les cumuls de véhicules entrants et sortants dans chaque tronçon du réseau. La formulation de ce modèle n'est pas explicite, ce qui justifie l'utilisation d'un filtre sans parfum. Les observations utilisées sont à la fois les boucles magnétiques à la sortie des tronçons et les véhicules traceurs après map-matching sur le réseau. L'algorithme est applicable en temps réel, sans connaissance a priori des affectations aux carrefours, et sous conditions éventuelles de perturbation. En sortie de filtre, on dispose des variances d'estimation, ce qui mesure la confiance dans le temps de parcours estimé et permet de rejeter des mesures aberrantes. La validation a été montrée sur la base de simulations sur un réseau simple à trois intersections, avec des flux variables et des perturbations, et pour quelques pourcents de véhicules traceurs<br>Travel time information is easy to be perceived by users and has the potential to reduce congestion on both temporal and spatial scales thanks to traveller assistance systems. This thesis addresses the problem of estimating the travel time in city, where traffic is characterized by variable flow and also significant perturbation from/to mid-link source/sink that are not equipped with traffic sensors. This flow disturbs the equation of conservation of vehicles. The bibliography contains deterministic and stochastic approaches, using, in specific experimental situations, either inductive loops or probe vehicles, or both. We propose a stochastic approach based on unscented Kalman filtering. Its dynamic model is based on the classical analytical procedure that considers the time difference between the total of incoming and outgoing vehicles for each segment of the network. The formulation of this model is not explicit, which justifies the use of an unscented filter. The observations are both loop counters at the exit sections and probe vehicles map-matched to the network. The algorithm is applicable in real time, without prior information on the percentage of turning movements at intersections, and under any condition of perturbation. The variance estimation output by the filter measures the confidence in the estimated travel time and allows the rejection of outliers. The validation was shown based on simulations on a simple network with three intersections, variable entering flows, variable percentage of perturbation, and a variable percentage of probe vehicles

Magical realism is a mode of literature that does not fit easily in anyone genre. In contrast, travel fiction has existed for so long that it has evolved into a straggling, cross-genre narrative form. Nevertheless, when coupled together, magical realist and travel fiction narratives share in common a number of features that allow the resulting text to traverse, create and challenge notions of space, territory and borders. I theorise that a number of key authors have already harnessed these features to criticise and deconstruct dominant hegemonic ideologies, including those of class, gender inequality and racial constructs. This thesis is an examination and discussion of how the two forms intertwine and complement each other, and how the features and characteristics encountered in one fonn augment and inform the other. I also posit clear examples of the phenomenon of intermingling in established texts. This thesis includes an in-depth discussion of key magical realist travel texts and the production of a new novel. The critical essay stretches to 27,000 words. The novel, a 71,000 word work, is written with these concerns as a central feature. It incorporates travel fiction and magical realist gemes in an attempt to update and re-deploy the form in a contemporary setting. The process of creating both texts necessitated a great deal of background reading from plimary and critical sources. My analysis of source texts and critical texts led to the conclusion that my initial theory was correct. Select examples of travel and magical realist literature employ narrative features to reinforce challenges to the established status-quo. However, the thesis further argues the two forms are interrelated in other ways, containing contradictory elements, a preoccupation with place and identity and the deconstruction and re-creation of myth. These features are also examined and discussed. Analytical reading and the writing of a related novel indicate that the use of magical realism and travel fiction results in a dynamic and effective challenge to the construction of oppressive societal norms.

The aim of this thesis is to write a multisensor navigation algorithm and to design a test setup. After doing these, test the algorithm by using the test setup. In navigation, dead reckoning is a procedure to calculate the position from initial position with some measured inputs. These measurements do not include absolute position data. Using only an inertial measurement unit is an example for dead reckoning navigation. Calculating position and velocity with the inertial measurement unit is highly erroneous because, this calculation requires integration of acceleration data. Integration means accumulation of errors as time goes. For example, a constant acceleration error of 0.1 m/s^2 on 1 m/s^2 of acceleration will lead to 10% of position error in only 5 seconds. In addition to this, wrong calculation of attitude is going to blow the accumulated position errors. However, solving the navigation equations while knowing the initial position and the IMU readings is possible, the IMU is not used solely in practice. In literature, there are studies about this topic and in these studies<br>some other sensors aid the navigation calculations. The aiding or fusion of sensors is accomplished via Kalman filter. In this thesis, a navigation algorithm and a sensor fusion algorithm were written. The sensor fusion algorithm is based on estimation of IMU errors by use of a Kalman filter. The design of Kalman filter is possible after deriving the mathematical model of error propagation of mechanization equations. For the sensor fusion, an IMU, two incremental encoders and a digital compass were utilized. The digital compass outputs the orientation data directly (without integration). In order to find the position, encoder data is calculated in dead reckoning sense. The sensor triplet aids the IMU which calculates position data by integrations. In order to mount these four sensors, an unmanned tracked vehicle prototype was manufactured. For data acquisition, an xPC&ndash<br>Target system was set. After planning the test procedure, the tests were performed. In the tests, different paths for different sensor fusion algorithms were experimented. The results were recorded in a computer and a number of figures were plotted in order to analyze the results. The results illustrate the benefit of sensor fusion and how much feedback sensor fusion is better than feed forward sensor fusion.

Thermonuclear fusion may become an attractive future power source. The most promising of all fusion machine concepts is the tokamak. Despite decades of active research, still huge tasks remain before a fusion power plant can go online. One of these important tasks deals with the interaction between the fusion plasma and the reactor wall. This work focuses on how eroded wall materials of different origin and mass are transported in a tokamak device. Element transport can be examined by injection of certain species of unique and predetermined origin, so called tracers. Tracer experiments were conducted at the TEXTOR tokamak before its final shutdown. This offered an unique opportunity for studies of the wall and other internal components: For the first time it was possible to completely dismantle such a machine and analyse every single part of reactor wall, obtaining a detailed pattern of material migration. Main focus of this work is on the high-Z metals tungsten and molybdenum, which were introduced by WF6 and MoF6 injection into the TEXTOR tokamak in several material migration experiments. It is shown that Mo and W migrate in a similar way around the tokamak and that Mo can be used as tracer for W transport. It is further shown how other materials - medium-Z (Ni), low-Z (N-15 and F), fuel species (D) - migrate and get deposited. Finally, the outcome of dust sampling studies is discussed. It is shown that dust appearance and composition depends on origin, formation conditions and that it can originate even from remote systems like the NBI system. Furthermore, metal splashes and droplets have been found, some of them clearly indicating boiling processes.<br><p>QC 20151203</p>

Controlled thermonuclear fusion may become an attractive future electrical power source. The most promising of all fusion machine concepts is called a tokamak. The fuel, a plasma made of deuterium and tritium, must be confined to enable the fusion process. It is also necessary to protect the wall of tokamaks from erosion by the hot plasma. To increase wall lifetime, the high-Z metal tungsten is foreseen as wall material in future fusion devices due to its very high melting point. This thesis focuses on the following consequences of plasma impact on a high-Z wall: (i) erosion, transport and deposition of high-Z wall materials; (ii) fuel retention in tokamak walls; (iii) long term effects of plasma impact on structural machine parts; (iv) dust production in tokamaks. An extensive study of wall components has been conducted with ion beam analysis after the final shutdown of the TEXTOR tokamak. This unique possibility offered by the shutdown combined with a tracer experiment led to the largest study of high-Z metal migration and fuel retention ever conducted. The most important results are:   - transport is greatly affected by drifts and flows in the plasma edge; - stepwise transport along wall surfaces takes place mainly in the toroidal direction; - fuel retention is highest on slightly retracted wall elements; - fuel retention is highly inhomogeneous.   A broad study on structural parts of a tokamak has been conducted on the TEXTOR liner. The plasma impact does neither degrade mechanical properties nor lead to fuel diffusion into the bulk after 26 years of duty time. Peeling deposition layers on the liner retain fuel in the order of 1g and represent a dust source. Only small amounts of dust are found in TEXTOR with overall low deuterium content. Security risks in future fusion devices due to dust explosions or fuel retention in dust are hence of lesser concern.<br><p>QC 20170630</p>

Depuis le début des années 2000, le développement de nouveaux traitements pour les troubles neurologiques et psychiatriques a significativement ralenti, notamment en raison des défis posés par la pénétration cérébrale des molécules, par leurs liaisons spécifiques aux cibles et par l’activation pharmacologique souhaitée. Ce contexte souligne la nécessité de nouvelles approches pour revitaliser les pipelines de médicaments neuropsychiatriques, en ciblant des composés plus spécifiques, efficaces et sûrs. La neuroimagerie se présente comme un outil intéressant pour répondre à ces enjeux, en facilitant l’identification de molécules innovantes et en approfondissant notre compréhension des mécanismes d’action des candidats-médicaments. Dans cette thèse, nous avons utilisé la neuroimagerie comme outil de développement pharmacologique à travers deux exemples de ligands sérotoninergiques. Le premier axe de cette thèse avait pour objectif de développer le premier radiotraceur TEP ciblant les récepteurs sérotoninergiques 5-HT2B afin d’aider à mieux comprendre la localisation et le rôle cérébral de ce récepteur peu exploré mais cible pharmacologique potentielle. Le second axe portait sur la caractérisation pharmacocinétique et pharmacodynamique du NLX-204, un candidat-médicament antidépresseur, agoniste biaisé des récepteurs 5-HT1A, grâce à l’association de deux modalités d’imagerie, la tomographie par émission de positons (TEP) et l’imagerie fonctionnelle ultrasonore (fUS). Afin de développer le premier radiotraceur des récepteurs 5-HT2B destiné à l’imagerie TEP, le RS-127445, outil pharmacologique connu pour sa haute affinité et sélectivité pour le récepteur 5-HT2B, a été radiomarqué au fluor 18 via un précurseur de radiosynthèse dérivé d’ylure d’iodonium. Les résultats des études d’imagerie préclinique in vivo et in vitro chez le rat ont montré que le [18F]RS-127445 présentait une bonne entrée initiale dans le cerveau du rat mais son fort taux de liaison non spécifique n’a pas permis de valider sa spécificité pour les récepteurs 5-HT2B. Dans un second projet, la combinaison innovante des imageries TEP et fUS en neuropharmacologie a permis une localisation précise de la liaison du NLX-204 dans le cerveau, tout en offrant une visualisation en temps réel de ses sites d’action. Le [18F]NLX-204 se lie majoritairement aux récepteurs 5-HT1A dans le tronc cérébral ainsi que les régions thalamiques et hypothalamiques, impliquées dans la coordination motrice et la douleur. Par ailleurs, l’imagerie fUS du NLX-204 chez l’animal éveillé, libre de ses mouvements, a révélé une activation dose-dépendante de régions corticales et de l’hippocampe, impliquées dans le traitement sensoriel, la mémoire et la coordination motrice. Ces résultats complémentaires apportent de nouvelles connaissances mécanistiques sur les agonistes biaisés des récepteurs 5-HT1A et soulignent le potentiel du NLX-204 comme candidat-médicament pour des traitements neurologiques ciblés<br>Since the early 2000s, the development of new treatments for neurological and psychiatric disorders has slowed significantly. This is partly due to the challenges posed by molecules' brain penetration, their specific binding to targets and the desired pharmacological activation. This context highlights the need for new approaches to revitalize neuropsychiatric drug pipelines, by targeting more specific, effective and safe compounds. Neuroimaging emerges as an interesting tool to address these challenges, facilitating the identification of innovative molecules and deepening our understanding of drug candidates' mechanisms of action. In this thesis, we employed neuroimaging as a tool for pharmacological development through two examples of serotoninergic receptors ligands. The aim of the first part of this thesis was to develop the first PET radiotracer targeting serotonin 5-HT2B receptors, to gain a better understanding of the location and cerebral function of this receptor, which remains to be extensively explored and has the potential to serve as a pharmacological target. The second area of research focused on the pharmacokinetic and pharmacodynamic characterisation of NLX-204, an antidepressant drug candidate and biased agonist of 5-HT1A receptors, using a combination of two imaging modalities, positron emission tomography (PET) and functional ultrasound imaging (fUS). In order to develop the first 5-HT2B receptor PET radiotracer, RS-127445, a pharmacological tool known for its high affinity and selectivity for the 5-HT2B receptor, was radiolabelled with fluorine 18 via a radiosynthesis from a iodonium ylide precursor. The results of preclinical in vivo and in vitro imaging studies in rats demonstrated that [18F]RS-127445 exhibited good initial entry into the rat brain. However, its high level of non-specific binding precluded validation of its specificity for 5-HT2B receptors in rats. In a second project, the innovative combination of PET and fUS imaging in neuropharmacology enabled precise localisation of NLX-204 binding in the brain, while providing real-time visualisation of its sites of action. [18F]NLX-204 binds predominantly to 5-HT1A receptors in the brainstem, thalamic and hypothalamic regions, known to be involved in motor coordination and pain. Furthermore, fUS imaging of NLX-204 in awake, freely moving animals revealed dose-dependent activation of cortical and hippocampal regions implicated in sensory processing, memory and motor coordination. These complementary results provide new mechanistic insights into biased 5-HT1A receptor agonists and underscore the potential of NLX-204 as a drug candidate for targeted neurological treatments

L'étude de la partie Nord-Est du plateau de Millevaches (Massif Central français) a permis de mettre en évidence treize faciès pétrographiques différents. Une caractérisation chimique de ces faciès par les éléments majeurs, les éléments en trace et les terres rares, a été entreprise. Les variations cristallochimiques des minéraux essentiels viennent conforter les discriminations établies par les études pétrographiques et géochimiques. La géochimie des éléments en trace montre que la fusion partielle est à l'origine de ces facies. Deux matériaux sources sont impliqués dans la génèse des facies à biotite et cordiérite ; l'ensemble des leucogranites (avec ou sans sillimanite) relève, par contre, de la fusion d'un matériau source unique. La diversité constatée, pétrographiquement et chimiquement, au sein de ce dernier ensemble est due à une ségrégation plus ou moins complète des phases résiduelles

Étude de différentes séries volcano-sédimentaires en zone pindique et sur des bordures (massifs du Lakmon, du Koziakas et du Vardoussia), en zone pelagonienne (ile d'Eubée), en zone Maliaque (Massif de l'Othrys) et dans le Dodécanèse (Iles de Tilos et de Rhodes). Dans chaque cas, un examen de l'environnement stratigraphique, ainsi qu'une étude pétrographique et géochimique des roches magmatiques permettent de définir les types de volcanisme suivant : calco-alcalin (massif d'Othrys, Eubée), transitionnel (massif d'Othrys, Eubée), alcalin (massif d'Othrys, Eubée, Dodécanèse). Plusieurs implications géodynamiques sont envisagées

Today, the main research field for the automotive industry is to find solutions for active safety. In order to perceive the surrounding environment, tracking nearby traffic objects plays an important role. Validation of the tracking performance is often done in staged traffic scenarios, where additional sensors, mounted on the vehicles, are used to obtain their true positions and velocities. The difficulty of evaluating the tracking performance complicates its development. An alternative approach studied in this thesis, is to record sequences and use non-causal algorithms, such as smoothing, instead of filtering to estimate the true target states. With this method, validation data for online, causal, target tracking algorithms can be obtained for all traffic scenarios without the need of extra sensors. We investigate how non-causal algorithms affects the target tracking performance using multiple sensors and dynamic models of different complexity. This is done to evaluate real-time methods against estimates obtained from non-causal filtering. Two different measurement units, a monocular camera and a LIDAR sensor, and two dynamic models are evaluated and compared using both causal and non-causal methods. The system is tested in two single object scenarios where ground truth is available and in three multi object scenarios without ground truth. Results from the two single object scenarios shows that tracking using only a monocular camera performs poorly since it is unable to measure the distance to objects. Here, a complementary LIDAR sensor improves the tracking performance significantly. The dynamic models are shown to have a small impact on the tracking performance, while the non-causal application gives a distinct improvement when tracking objects at large distances. Since the sequence can be reversed, the non-causal estimates are propagated from more certain states when the target is closer to the ego vehicle. For multiple object tracking, we find that correct associations between measurements and tracks are crucial for improving the tracking performance with non-causal algorithms.

Les termes acides et basiques des massifs ophiolitiques du Montgenèvre et de Haute- Ubaye sont étudiés en parallèle dans ce travail. L'étude pétrographique (notamment l'étude typologique des zircons pour les albitites) et la géochimie des éléments majeurs confirment la nature tholéilique des magmasétudiés. L'étude du comportement des éléments traces (Zr, Y, Nb,Th, Cr, REE) permet de mettre en évidence, pour le Montgenèvre,une évolution continue des roches basiques conduisant de basaltes enrichis en LREE, situés à la base de la série, à des basaltes appauvris en LREE (N -MORB typiques) tardifs et représenlés par les coulées aphyriques supérieures et certains filons doléritobasaltiques. Une telle évolution paraît refléter une fusion partielle dynamique, seule capable d'expliquer les variations importantes du rapport La/Sm. Les roches basiques de Haute- Ubaye montrent une plus grande homogénéité géochimique que celles du Montgenèvre avec une majorité de spectres de terres rares bombés. Les variations du taux de fusion partielle (Montgenèvre) et les caractères de la cristallisation fractionnée (Montgenèvre el Haute Ubaye) sont précisés. L'étude des éléments-lraces des albitites des deux ensembles révèle également la diversité géochimique de ces roches et vraissemblablement l'existence de plusieurs souches parentales. Au Montgenèvre, seules les albitites massives de la Replatte du Gondran (Chenaillet) peuvent être considérées comme cogénétiques des basaltes en coussins. Les albitites de la Punta Rascia (Italie) seraient plus anciennes, postérieures au premier épisode magmatique à l'origine de la grande masse gabbroïque du Montgenèvre. En Haute- Ubaye, il n'existe aucune filiation géochimique nette entre les roches basiques et les roches acides. Les observations de terrain et les données géochimiques permettent d'envisager une évolution temporelle complexe où les termes effusifs paraissent s'être mis en place tardivement et latéralement par rapport au rift océanique. L'étude typologique des zircons des éléments acides d'une brèche polygénique située à l'interface ophiolite-sédiments dans la partie orientale du massif a montré la nature calco-alcaline de certains de ces éléments. Par comparaison avec des brèches semblables étudiées dans le Queyras, ces éléments pourraient provenir du démantèlement de la croûte continentale voisine. Ainsi, les ophiolites du Montgenèvre sont, par leur position structurale et leurs caractéristiques géochimiques, à rapporter vraissemblablement aux premiers termes du magmatisme océanique téthysien.