L’un des objectifs de la microélectronique est de concevoir et fabriquer des SoCs de petites tailles, à moindre coût et visant des marchés tel que l’internet des objets. À matériel fixe sur lequel l’on ne dispose d’aucune marge de manœuvre, l’un des challenges pour un développeur de logiciels embarqués est d’écrire son programme de manière à ce qu’à l’exécution, le logiciel développé puisse utiliser au mieux les capacités de ces SoCs. Cependant, ces programmes n’utilisent pas toujours correctement les capacités de traitement disponibles sur le SoC. L’estimation et l’optimisation de la performance du logiciel devient donc une activité cruciale. A l’exécution, ces programmes sont très souvent victimes de l’apparition de cycles de gel de processeur dus à l’absence de données en mémoire cache. Il existe plusieurs approches permettant d’éviter ces cycles de gel de processeur. Par l’exemple l’utilisation des options de compilation adéquates pour la génération du meilleur code exécutable possible. Cependant les compilateurs n’ont qu’une idée abstraite (sous forme de formules analytiques) de l’architecture du matériel sur lequel le logiciel s’exécutera. Une alternative est l’utilisation des processeurs « Out–Of–Order ». Mais ces processeurs sont très couteux en terme de coût de fabrication car nécessites une surface de silicium importante pour l’implantation de ces mécanismes. Dans cette thèse, nous proposons une méthode itérative basée sur les plateformes virtuelles précises au niveau du cycle qui permet d’identifier les instructions du programme à optimiser responsables à l’exécution, de l’apparition des cycles de gel de processeur dus à l’absence de données dans le cache L1. L’objectif est de fournir au développeur des indices sur les emplacements du code source de son programme en langage de haut niveau (C/C++ typiquement) qui sont responsables de ces gels. Pour chacune de ces instructions, nous fournissons leur contribution au rallongement du temps d’exécution totale du programme. Finalement nous estimons le gain potentiel maximal qu’il est possible d’obtenir si tous les cycles de gel identifiés sont évités en insérant manuellement dans le code source du programme à optimiser, des instructions de pré–chargement de données dirigé par le logiciel. / One of microelectronics purposes is to design and manufacture small-sized, low-cost SoCs targeting markets such as the Internet of Things. With fixed hardware on which there is no possible flexibility, one of the challenges for an embedded software developer is to write his program so that, at runtime, the software developed can make the best use of these SoC capabilities. However, these programs do not always properly use the available SoC processing capabilities. Software performance estimation and optimization is then a crucial activity. At runtime, these programs are very often victims of processor data stall cycles. There are several approaches to avoiding these processor data stall cycles. For example, using the appropriate compilation options to generate the best executable code. However, the compilers have only an abstract knowledge (as analytical formulas) of the hardware architecture on which the software will be executed. Another way of solving this issue is to use Out-Of- Order processors. But these processors are very expensive in terms of manufacturing cost because they require a large silicon surface for the implementation of the Out-Of-Order mechanism. In this thesis, we propose an iterative methodology based on cycle accurate virtual platforms, which helps identifying precisely instructions of the program which are responsible of the generation of processor data stall cycles. The goal is to provide the developer with clues on the source code lignes of his program’s in high level language (C/C++ typically) which are responsible of these stalls. For each instructions, we provide their contribution to lengthening of the total program execution time. Finally, we estimate the maximum potential gain that can be achieved if all identified stall cycles are avoided by manually inserting software preloading instructions into the source code of the program to optimize.

No description available.

No description available.

Modern systems fulfil calculation tasks across the hardware- software boundary. Tasks are divided into coarse parallel subtasks that run on distributed resources. These resources are classified into a software (SW) and a hardware (HW) domain. The software domain usually contains processors for general purpose or digital signal calculations. Dedicated co-processors such as encryption or video en-/decoding units belong to the hardware domain. Nowadays, a decision in which domain a certain subtask will be executed in a system is usually taken during system level design. This is done on the basis of certain assumptions about the system requirements that might not hold at runtime. The HW/SW partitioning is static and cannot adapt to dynamically changing system requirements at runtime. Our contribution to tackle this, is to combine a ESL based HW/SW codesign methodology with a coarse grained reconfigurable System on Chip architecture. We propose this as Preemptive HW/SW-Threading.

Viti Levu, the main island of Fiji, is located in a seismically active area within the Fiji Platform - a remnant island arc that lies in a diffuse plate boundary zone between the Pacific and Australian tectonic plates in the southwest Pacific. The southeast coast of Viti Levu is a highly developed and populated part of Fiji and is vulnerable to the effects of large earthquakes that are expected to occur both onshore and offshore. The structural framework and the origin of seismicity within the Fiji Platform, as well as the seismic and tsunami hazards of central and southeast Viti Levu are investigated. The upper crust of southeast Viti Levu is dissected by several intersecting fault/lineament zones. These are mapped from remote sensing imagery of the surface (topography, radar, and aerial photos) and of the basement (magnetic), and have been subject to rigorous statistical tests of reproducibility and verification with field mapped fault data. Lineaments on the various imagery correlate with faults mapped in the field and show spatial continuity between and beyond mapped faults, thereby providing a fuller coverage of regional structural patterns than previously known. Some fault/lineament zones extend beyond the coastline to the offshore area of southeast Viti Levu. Here high resolution SeaBAT 8160 multibeam bathymetry data and seismic reflection data show that the fault zones occur along, and exert control on the locations of a number of linear submarine canyons. The morpho-structural expression of these canyons are contiguous with fault controlled physiographic features mapped on the nearshore marginal shelf (rectilinear bays and peninsulas, reef passages) and on land (fault valleys, slope and drainage alignments forming lineaments). The canyons are considered to have developed from several cycles of downslope incising and infilling events, whilst their positions were still primarily controlled by zones of weakness created by the fault zones. The principal fault sets in southeast Viti Levu represent generations of regional tectonic faulting that pervaded the Fiji Platform during and after disruption of the proto Fijian arc in the Middle to Late Miocene. These fault sets combine to form a complex network of interlocking faults creating a fault mesh that divides the upper crust into a number of fault blocks ranging from ~2 to 30 km. It is inferred that the fault mesh evolved throughout the Neogene as a response to the anticlockwise rotation of the Fiji Platform through progressive development of different fault sets and intervening crustal block rotations. Regional tectonic deformation is presently accommodated in a distributed manner through the entire fault mesh. Low magnitude earthquakes (<M4) occur regularly and may represent ruptures along short linking segments of the fault mesh, while infrequent larger earthquakes (>M4) may result from complex rupture propagation through several linking fault segments of the mesh that lie close to optimum stress orientations. This interpreted model of distributed deformation through the fault mesh for southeast Viti Levu is inferred to be characteristic of the style of active deformation that occurs throughout the entire Fiji Platform. Seismic activity is primarily responsible for triggering submarine landslides that occur on the southeastern slope of Viti Levu. These slides typically occur on the outer barrier reef edge, as well as in submarine canyon heads and walls, and in the mid slope areas. They are characteristically translational and lack bathymetric evidence for displaced masses. Morphometric analysis and empirical modelling, show that slides triggered at shallow water depths, within 5 km of the coastline, at the outer barrier reef edge and submarine canyon heads, produce the largest near-field tsunami amplitudes. Such slides are interpreted to represent a significant local tsunami hazard. A detailed case study of the destructive 1953 Suva tsunami that followed the Ms 6.75 Suva earthquake, reveals that the source of this tsunami was a 60 million cubic metre submarine landslide at the head of the Suva Canyon, 4 km to the WSW of Suva City. A test simulation of this tsunami using the Geowave tsunami generation, propagation and inundation model, closely replicates the wave heights and arrival times recorded in 1953. This simulation also reveals that high variability in tsunami impact over short coastal distances of southeast Viti Levu is attributable to the complex interplay of wave propagation with the barrier reef system, erratic lagoon bathymetry and the irregularly shaped coastline. A predictive simulation using Geowave, based on an incipient failure in the 1953 source area and on a potentially worse case scenario event at or near high-tide, is used to show a maximum vertical run up of at least 4 m and a maximum horizontal inundation level of at least 400 m at the Suva coast. The seismic hazard of five sites on Viti Levu, including Suva City, Navua and Nausori Towns, and the Monsavu and Nadarivatu dam sites, is evaluated using a deterministic approach, and seven newly identified crustal fault earthquake source structures. The maximum magnitudes interpreted for these structures, estimated using empirical relationships, range from Mw 6.8 to 7.6. The Suva Canyon Fault, the Naqara Fault, the Mavuvu/Fault Lineament Zone and the Nasivi Fault provide the controlling maximum credible earthquakes (CMCE) at all the five sites. The CMCE peak ground acceleration values for Suva City range from 0.4g to 0.6g, for Nausori Town from 0.18g to 0.2g, for Navua Town from 0.27g to 0.32g, for Monasavu from 0.39g to 0.42g, and for Nadarivatu from 0.23g to 0.33g. The horizontal spectral accelerations at a period equal to 0.2 seconds, calculated using the CMCEs, are comparable to accelerations derived by probabilistic methods that have return periods between 50 and over 1000 years.

Made available in DSpace on 2016-06-02T19:05:47Z (GMT). No. of bitstreams: 1 3388.pdf: 1949798 bytes, checksum: 1a2073d52b18588ac8cb15ca66beb5d9 (MD5) Previous issue date: 2008-05-22 / Universidade Federal de Sao Carlos / Background: The Personal Software ProcessTM (PSP) is a methodology that can be used by software developers to improve the predictability, the productivity and the quality of their personal work. However, the great number of activities and metrics involved in the PSP makes its application very laborious and difficult. Several PSP supporting tools have been proposed in order to make its application more feasible, and such tools evolved as more experiments were conducted with the PSP. There are now three generations of PSP supporting tools, whereas third generation supporting tools are characterized by being able to collect and analyze Software Engineering metrics automatically and unobtrusively. Objective: This study s objective is to determine the difficulties and benefits of using third generation tools to support the PSP application. Method: The GQM paradigm was used to set up two Case Studies, one being in the academic environment with undergraduate students and another in a small software development company. Both case studies featured the PSP application supported by third generation tools. Results: Results show that third generation supporting tools have some advantages over previous generations tools, especially in the coding phase. However, third generation tools have limited PSP activities support and introduce new adoption problems. Conclusion: For a successful long-term PSP application, it is necessary that the supporting tools are completely adapted and integrated in the developers work environment, with both second and third generations tools characteristics. / Cenário: O Personal Software ProcessTM (PSP) é uma metodologia que pode ser usada por desenvolvedores de software para melhorar a previsibilidade, a produtividade e a qualidade de seu trabalho pessoal. No entanto, a grande quantidade de atividades e métricas envolvidas no PSP torna a sua aplicação bastante trabalhosa e difícil. Várias ferramentas de apoio foram propostas com o intuito de facilitar a aplicação da metodologia, sendo que tais ferramentas evoluíram à medida que estudos sobre a aplicação do PSP eram realizados. Assim, existem hoje três gerações de ferramentas de apoio ao PSP, sendo que as ferramentas de 3ª geração caracterizam-se por coletar e analisar métricas de Engenharia de Software automaticamente, sem a necessidade de intervenção direta do desenvolvedor. Objetivo: O objetivo deste trabalho é determinar as dificuldades e os benefícios do uso de ferramentas de 3ª geração para o apoio à aplicação do PSP. Método: Utilizando o paradigma GQM, foram elaborados dois estudos de caso, sendo um em ambiente acadêmico e outro em uma pequena empresa de desenvolvimento de software. Ambos os estudos de caso envolveram a aplicação do PSP com o apoio de ferramentas de 3ª geração. Resultados: Os resultados mostram que ferramentas de 3ª geração apresentam algumas vantagens em relação às ferramentas de gerações anteriores, especialmente na fase de codificação. No entanto, tais ferramentas apóiam um número limitado de atividades do PSP, além de introduzirem novos problemas de adoção. Conclusão: Para que a aplicação do PSP tenha sucesso em longo prazo é necessário que a ferramenta de apoio seja completamente adaptada e integrada ao ambiente de trabalho do desenvolvedor, apresentando tanto características de ferramentas de 2ª quanto de 3ª gerações.

Les véhicules autonomes, tels que les drones, sont utilisés dans différents domaines d'application pour exécuter des missions simples ou complexes. D’un côté, ils opèrent généralement dans des conditions environnementales incertaines, pouvant conduire à des conséquences désastreuses pour l'humain et l'environnement. Il est donc nécessaire de surveiller continuellement l’état de santé du système afin de pouvoir détecter et localiser les défaillances, et prendre la décision en temps réel. Cette décision doit maximiser les capacités à répondre aux objectifs de la mission, tout en maintenant les exigences de sécurité. D’un autre côté, ils sont amenés à exécuter des tâches avec des demandes de calcul important sous contraintes de performance. Il est donc nécessaire de penser aux accélérateurs matériels dédiés pour décharger le processeur et répondre aux exigences de la rapidité de calcul.C’est ce que nous cherchons à démontrer dans cette thèse à double objectif. Le premier objectif consiste à définir un modèle pour l’état de santé et la décision. Pour cela, nous utilisons les réseaux Bayésiens, qui sont des modèles graphiques probabilistes efficaces pour le diagnostic et la décision sous incertitude. Nous avons proposé un modèle générique en nous basant sur une analyse de défaillance de type FMEA (Analyse des Modes de Défaillance et de leurs Effets). Cette analyse prend en compte les différentes observations sur les capteurs moniteurs et contextes d’apparition des erreurs. Le deuxième objectif était la conception et la réalisation d’accélérateurs matériels des réseaux Bayésiens d’une manière générale et plus particulièrement de nos modèles d’état de santé et de décision. N’ayant pas d’outil pour l’implémentation embarqué du calcul par réseaux Bayésiens, nous proposons tout un atelier logiciel, allant d’un réseau Bayésien graphique ou textuel jusqu’à la génération du bitstream prêt pour l’implémentation logicielle ou matérielle sur FPGA. Finalement, nous testons et validons nos implémentations sur la ZedBoard de Xilinx, incorporant un processeur ARM Cortex-A9 et un FPGA. / Autonomous vehicles, such as drones, are used in different application areas to perform simple or complex missions. On one hand, they generally operate in uncertain environmental conditions, which can lead to disastrous consequences for humans and the environment. Therefore, it is necessary to continuously monitor the health of the system in order to detect and locate failures and to be able to make the decision in real time. This decision must maximize the ability to meet the mission objectives while maintaining the security requirements. On the other hand, they are required to perform tasks with large computation demands and performance requirements. Therefore, it is necessary to think of dedicated hardware accelerators to unload the processor and to meet the requirements of a computational speed-up.This is what we tried to demonstrate in this dual objective thesis. The first objective is to define a model for the health management and decision making. To this end, we used Bayesian networks, which are efficient probabilistic graphical models for diagnosis and decision-making under uncertainty. We propose a generic model based on an FMEA (Failure Modes and Effects Analysis). This analysis takes into account the different observations on the monitors and the appearance contexts. The second objective is the design and realization of hardware accelerators for Bayesian networks in general and more particularly for our models of health management and decision-making. Having no tool for the embedded implementation of computation by Bayesian networks, we propose a software workbench covering graphical or textual Bayesian networks up to the generation of the bitstream ready for the software or hardware implementation on FPGA. Finally, we test and validate our implementations on the Xilinx ZedBoard, incorporating an ARM Cortex-A9 processor and an FPGA.

Este trabalho caracteriza a evolucao tectónica, identificando a cronologia dos principais eventos tectono-magmáticos do Pré-Cambriano Boliviano. A complexa evolucao geológica do Oriente da Bolívia se estende desde o Paleo a Mesoproterozoico compreendendo as provincias Rio Negro Juruena, Rondoniana San Ignacio e Sunsás na regiao conhecida como Bloco Paragua. Diversos métodos de estudo foram adotados na pesquisa tendo em vista tratar-se de um terreno com evolução policíclica e incluiram, alem do mapeamento geológico e petrografía dos principais tipos de rocha, a metodologia U-Pb para determinação da idade de corpos graníticos e a metodologia Sm-Nd na estimativa de idade das fontes destes corpos plutônicos e inferências de ordem petrogenética, bem como dados geoquímicos obtidos para detalhamento das interpretações petrogenéticas. Nas interpretações houve ainda a avaliação critica da literatura recente, a integração de dados de campo, aeromagnéticos e aero-radiométricos, inclusive embasadas na experiência profissional do autor. Os dados obtidos na última década modificaram substancialmente a concepcao do Pré-Cambriano Boliviano, tendo sido caracterizados tres conjuntos litológicos temporalmente distintos antecedendo a orogenia San Ignacio. O granito Correreca na parte meridional da area possui idade 207Pb-206Pb de 1,92 1,89 Ga, com modelo de idades TDM de 2,8 a 2,9 Ga e valores de Nd(t) de -8,5 e -9,4. A Suite Yarituses composta pelos granitos La Cruz, Refugio e San Pablo possui quimismo calcio-alcalino. Os dados U-Pb SHRIMP, TIMS e abrasão por laser-ICPMS indicam a formação desta suíte no lapso temporal entre 1673 a 1621 Ma. A idade de cristalização U-Pb SHRIMP do granito La Cruz é de 1673 ± 21 Ma, idade modelo TDM de 1,83 Ga e valor de Nd(t) de + 2.1 indicativo de derivação mantélica. O granito Refugio tem idade U-Pb TIMS de 1673 ± 25 Ma e o pluton San Pablo idade ICPMS por laser ablasion de 1621 ± 80 Ma (idade TDM de 1,7 Ga e valor de Nd(t) de + 3,5). Este conjunto de dados sugere uma derivação mantelica principal para a suite Yarituses. O granodiorito San Ramón possui uma idade de cristalização de 1429 ± 4 Ma (SHRIMP), TDM de 1,7 Ga, e Hf(t) entre + 3,49 e +5,47 e representa um evento de geração da crosta, a partir de material juvenil. O magmatismo, deformação e metamorfismo da orogênese San Ignácio constitui o principal evento representado na área de estudo, cujo maior representante é o Complexo Granitoide Pensamiento com seus plutons sin a tardi-cinemáticos e tardi a pos-cinemáticos. Os granitos San Martín, La Junta e Diamantina possuem idades de cristalizacao de 1373- 1340 Ma, idades modelo TDM de 1,6 a 2,0 Ga, com valores de Nd(t) de + 2.0 ate -4,0. Os granitos Las Maras, Talcoso, Limonal e San Andrés produziram idades de cristalização de 1347 a 1275 Ma. As idades TDM dos granitos Limonal e San Andrés correspondem a 1,9 e 1,8 e Nd(t) de -1,4 e 1,6 respectivamente. A geoquímica em rocha total indica uma composição compatível com arco magmático, corroborando a assinatura acima dos parâmetros petrogeneticos. Em suma, a orogênese San Ignácio representa um arco acrescionário de natureza continental que construiu a arquitetura final da província Rondoniana-San Ignacio pela colisão entre o Bloco Paraguá e a província Rio Negro-Juruena. A evolução mesoproterozoica finaliza com a formação da faixa colisional Sunsás. Esta orogênese produziu plutonismo sin a tardi cinematico e tardi a cinemático marcando o limite com o bloco Paragua. A natureza alóctone e colisional do orogeno Sunsás como o evento mais jovem do Cráton Amazônico é marcada por frentes tectônicos, bem definidos de sentido sinistral, convergentes para o Bloco Paragua. / This work characterizes the tectonic and magmatic evolution of the Precambrian shield of Bolivia. The complex geological evolution of the eastern Bolivia extends from the Paleo- to Mesoproterozoic, and can be related with the magmatic and metamorphic events that are ascribed to the Rio Negro - Juruena (1.78-1.60 Ga), Rondonian - San Ignacio (1.56-1.30 Ga) and Sunsás Aguapei (1.25-1.00 Ga) provinces, known in Bolivia as the Paragua block. Several methods of study were adopted in the research with the scope that this is a land with polycyclic evolution. As such our study included, besides the geological mapping and petrography of major rock types, the U-Pb age determinations of granitoid rocks, Sm-Nd and Rb-Sr isotopic analyses, as well as geochemical data. At the interpretation there was the critical evaluation of recent papers, the integration of field data, aeromagnetic and aero-radiometric, including the field experience of the author. The data obtained in the last decade have substantially changed the geology of the Bolivian Precambrian shield. It has been characterized three temporally distinct granite suites preceding the San Ignacio orogeny (1.37-1.30 Ga): the Correreca granite in the southern part of the area has 207Pb/206Pb age from 1.92 to 1.89 Ga, with TDM model ages of 2.8 to 2.9 Ga and values of Nd(t) of -8.5 and -9.4; the Yarituses suite (La Cruz, Refugio and San Pablo granites) shows calc-alkaline signature. Data U-Pb SHRIMP, TIMS and ICPMS laser ablation indicate the formation of this suite between 1673 to 1621 Ma. The U-Pb SHRIMP crystallization age of La Cruz granite is 1673 ± 21 Ma, TDM model age of 1.83 Ga and Nd(t) of +2.1 indicative of a predominantly mantle source. The Refugio granite has U-Pb TIMS age of 1673 ± 25 Ma and the San Pablo pluton yields a ICPMS Laser ablation age of 1621 ± 80 Ma (TDM age of 1.7 Ga and Nd(t) +3.5). These data suggest again a mantle source for the Yarituses suite. The San Ramon granodiorite event has a crystallization age of 1429 ± 4 Ma (SHRIMP), TDM of 1.7 Ga, and Hf(t) between +3.49 and +5.47 and represents a juvenile accreted episode. The magmatism, deformation and metamorphism of San Ignacio orogeny is the main event of the study area, represented by the Pensamiento Granitoid Complex with sin to late-kinematic and late to post-kinematic plutons. The San Martín, La Junta and Diamantina granites have crystallization ages of 1373 - 1340 Ma, TDM model ages from 1.6 to 2.0 Ga, with values of Nd(t) from 2.0 up to -4.0. The Las Maras, Talcoso, Limonal and San Andrés granites yielded crystallization ages of 1347-1275 Ma. The TDM ages of Limonal and San Andrés granites are between 1.9 and 1.8 Ga and the Nd(t) values of -1.4 and +1.6 respectively. The whole rock geochemistry of these granites indicates a composition consistent with the magmatic arc. Thus the San Ignacio orogeny represents a continental accretionary arc that built the final architecture of the Rondonian-San Ignacio province (1.56-1.30 Ga) by the collision between the Paragua block and the Rio Negro -Juruena province (1.78-1.60 Ga). The Mesoproterozoic evolution of the SW margin of the Amazonian craton ends with the formation of the Sunsás collisional belt that produced sin to-late and late topost- kinematic plutonism. The allochthonous and collisional nature of the Sunsás orogeny is marked by tectonic fronts, with well-defined sinistral sense, converging towards the Paragua block.

Orientador: Osvair Vidal Trevisan / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecanica e Instituto de Geociencias / Made available in DSpace on 2018-08-12T11:42:33Z (GMT). No. of bitstreams: 1 Moreira_RichardDouglasRibeiro_M.pdf: 5216671 bytes, checksum: f2bb159b2f1d91175972a75a6c43e44c (MD5) Previous issue date: 2006 / Resumo: A injeção de vapor auxiliada por drenagem gravitacional em poço único, denominada SWSAGD (Single Well Steam Assisted Gravity Drainage), é um processo de recuperação terciária desenvolvido com um único poço horizontal. Foram estudadas diversas estratégias, através de simulação numérica, visando a aplicação desta técnica com dados pertinentes a um campo da bacia do Espírito Santo. As estratégias têm diferentes opções como a injeção cíclica prévia e a colocação de obturadores no poço produtor. O desempenho do processo de recuperação para as diferentes estratégias é comparado sempre com aquele obtido para o processo do Dual Well - SAGD para as mesmas condições. São feitas também comparações com a produção primária por poço horizontal e entre as diversas estratégias geradas. A influência de alguns parâmetros - comprimento e posição entre poços, zona de injeção e produção - são apresentados. Com todas estas estratégias de melhoria para o processo SW-SAGD, alcança-se um processo com recuperação maior que os resultados decorrentes do SAGD tradicional com dois poços. / Abstract: The Single Well Steam Assisted Gravity Drainage (SW-SAGD) is a tertiary recovery process developed with an single horizontal well. The objective of this research is to study, with numerical simulation, the application of the SW-SAGD technique to a field dates located in the Espírito Santo Basin. Several strategies were studied for this process using previous cyclic injection and packers. The strategies improved the horizontal well production and enhanced the oil recovery. Comparisons are made along the study between the performance of oil recovery for the developed strategies and the performance of the DW-SAGD at the same operating and field conditions. Comparisons with the primary recovery using horizontal wells and between the strategies were used to improve and choose the best options. The influence of some parameters - length and position between wells, injection and production zones - are presented. As a result of all the improvement, a new strategy for the SW-SAGD process is reached, providing an oil recovery higher than from the DW-SAGD. / Mestrado / Reservatórios e Gestão / Mestre em Ciências e Engenharia de Petróleo

L’archipel volcanique actif du Vanuatu s’édifie au coeur du pacifique sud-ouest au niveau de la frontière convergente des plaques australienne et pacifique. Je présente ici une nouvelle étude géochimique des laves du Vanuatu à partir de la détermination des compositions en éléments majeurs et traces, et des compositions isotopiques (Sr-Nd-Hf-Pb) d’une centaine d’échantillons de laves (< 2 Ma).L’étude des magmas les plus primitifs a permis de mettre en évidence la variation de composition des sources mantelliques le long de l’arc et d’individualiser 3 segments : "central", dans la zone de collision de la ride D’Entrecasteaux, "sud" en face du bassin Nord Fidjien, et "extrême sud" en face du bassin Sud Fidjien. La composition des roches des différentes structures subductées influence celles des laves des volcans adjacents via le composant de subduction, sous forme de fluides et de produits de fusion. Les laves des îles situées en face de la ride D’Entrecasteaux sont issues d’un manteau enrichi ("type-MORB indien"), différent de celui échantillonné par les autres laves ("type-MORB pacifique"). Cette ride apporte probablement en subduction un composant ancien, pouvant être assimilé à un fragment de croûte inférieure.L’étude locale de certaines îles a permis de caractériser la différenciation des laves par cristallisation fractionnée, d’identifier des processus d’assimilation crustale, et de révéler la présence de magma provenant de portions de manteau distinctes, ayant subi un métasomatisme différent.Ces travaux révèlent une extrême hétérogénéité du manteau sous l’arc du Vanuatu, témoignant de la complexité des processus géologiques impliqués au niveau de cette zone de subduction. / The Vanuatu island arc is located in the SW Pacific at the convergent boundary between the Australian and Pacific plates. I present here a new geochemical study of the Vanuatu lavas based on major and trace element as well as isotopic (Sr-Nd-Hf-Pb) analyses of approximately one hundred lava samples (< 2 My).Examination of the most primitive magmas reveals the variation of the mantle source composition along the arc and demarcates three segments: a "central" segment in front the d’Entrecasteaux ridge, a "south" segment in front of the North Fiji Basin, and a "far south" segment in front of the South Fiji Basin. The composition of the different subducted basins highly influenced those of the adjacent volcanoes via mass transfer of aqueous fluid and melt subduction components. Lavas from islands facing the D'Entrecasteaux Ridge come from an enriched mantle (Indian MORB-like), distinct from that sampled at other islands (Pacific MORB-like). Additionally, this ridge likely carries with it an ancient component that may reflect entrainment of lower crust material.Localized study of some volcanic islands reveals processes related to the differentiation of lavas by fractional crystallization, processes related to crustal assimilation, and indicates the presence of magma coming from different portions of mantle with slightly different metasomatic histories.This body of work highlights extreme mantle heterogeneity under Vanuatu islands arc and testifies to the complex nature of its associated geological processes.