Made available in DSpace on 2016-12-12T20:27:38Z (GMT). No. of bitstreams: 1 Camila Bianka Silva Bastos.pdf: 1963598 bytes, checksum: bee88eacc3f6e3c327425297316a691d (MD5) Previous issue date: 2015-02-27 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / This dissertation presents a study of the operation of two different three-phase grid-connected test-grids with the connection of a 1MWp photovoltaic system. Two analysis methods are used to evaluate the impacts of this photovoltaic systeM, these methods being conventional static analysis and the analysis known as Quasi-Static Time-Series Analysis. Despite the fact that all grids have unique characteristics, it is important to use test-grids, which simulate the real grid characteristics, to analyze the kinds of problems that can occur and then look for alternatives, if necessary. The impacts evaluated are related to the system losses, minimized with the allocation study of the generation on the grid, voltage profile and tap position curve, when automatic load tap changers are used. It was verified that the photovoltaic system interconnection point is the most influenced one after its connection to the grid. The Quasi-Static Time-Series Analysis allow the correct evaluation of the load-generation interaction, running the time series power flow through estimated data for the load and irradiance curves during 168 hours. The conventional static analysis only considers critical operation conditions, like minimum and maximum load, and no generation or maximum generation, and does not evaluate different case scenarios that occur in reality. The photovoltaic systems can bring many advantages to the electric systems, like the improvement on the final consumer voltage profile, line losses reduction, and also environmental impacts reduction. However, with the increase of distributed photovoltaic generation on the electrical grid, it s necessary to be aware of the impacts that this may cause by performing interconnection studies. / Esta dissertação apresenta um estudo da operação de uma rede teste trifásica de média tensão com a interligação de um sistema fotovoltaico de 1,0 MWp. Dois métodos de análise são utilizados para avaliar os impactos deste sistema fotovoltaico, sendo estes métodos as análises estáticas convencionais eas análises conhecidas como Quasi-Static Time-Series Analysis. Apesar de cada rede elétrica apresentar características únicas, é importante a utilização de sistemas testes, que simulam as características de sistemas reais, para analisar que tipos problemas podem surgir e então buscar alternativas, se necessário. Os impactos avaliados se referem às perdas no sistema, minimizadas com a correta alocação da geração, perfil de tensão e curva de posição do tap, no caso de transformador com comutação automática de tap. Contata-se que o ponto de conexão do sistema fotovoltaico é o mais influenciado pela sua conexão à rede. As análises QSTS possibilitam avaliar corretamente a iteração entre carga e geração, efetuando o fluxo de potência consecutivo através de dados estimados para as curvas de carga e de irradiância solar ao longo de 168 horas. Já as análises convencionais consideram apenas condições críticas de operação, como por exemplo, carga leve ou nominal e geração nula ou máxima, não avaliando então diferentes cenários de operação que ocorrem na prática. Os sistemas fotovoltaicos podem trazer muitos benefícios aos sistemas elétricos, como melhoria do perfil de tensão de atendimento ao consumidor, redução de perdas nas linhas, além da redução nos impactos ambientais. Entretanto, com o aumento de geração fotovoltaica distribuída na rede, é necessário estar atento aos impactos que isto pode causar através de estudos de interconexão.

Ce travail de thèse concerne la mécanique numérique du contact entre solides déformables. Il s'agit de contribuer à l'amélioration de la performance, de la fiabilité et de la robustesse des algorithmes et des modèles numériques utilisés dans les codes éléments finis en particulier Code_Aster qui est un code libre développé par Électricité De France (EDF) pour ses besoins en ingénierie. L'objectif final est de traiter les problèmes industriels de grande dimension avec un temps de calcul optimisé. Pour parvenir à ces objectifs, les algorithmes et formulations doivent prendre en compte les difficultés liées à la mécanique non régulière à cause des lois de Signorini-Coulomb ainsi que la gestion des non linéarités dûes aux grandes déformations et aux comportements des matériaux étudiés.Le premier axe de ce travail est dédié à une meilleure compréhension de la formulation dite de « Lagrangien stabilisé » initialement implémentée dans le code. Il a été démontré l'équivalence entre cette formulation et la formulation bien connue de « Lagrangien augmenté ». Les caractéristiques mathématiques liées aux opérateurs discrets ont été précisées et une écriture énergétique globale a été trouvée. / This work deals with computational contact mechanics between deformable solids. The aim of this work is to improve the performance, the reliability and the robustness of the algorithms and numerical models set in Code_Aster which is finite element code developped by Électricité De France (EDF) for its engineering needs. The proposed algorithms are used to solve high dimensional industrial problems in order to optimize the computational running times. Several solutions techniques are available in the field of computational contact mechanics but they must take into account the difficulties coming from non-smooth aspects due to Signorini-Coulomb laws coupled to large deformations of bodies and material non linearities. Firstly the augmented Lagrangian formulation so-called « stabilized Lagrangian » is introduced. Successively, the mathematical properties of the discrete operators are highlighted and furthermore a novel energetic function is presented. Secondly the kinematical condition with regard to the normal unknowns are reinforced through unconstrained optimization techniques which result to a novel formulation which is so-called « non standard augmented Lagrangian formulation ». Three types of strategies are implemented in the code. The generalized Newton method is developped : it is a method in which all the non linearities are solved in one loop of iterations. The partial Newton method is an hybrid technique between the generalized Newton one and a fixed point method.

La mousse de polyuréthane est un matériau cellulaire caractérisé par un spectre de propriétés mécaniques intéressant : une faible densité, une capacité à absorber l’énergie de déformation et une faible raideur.Elle présente également des propriétés telles qu’une excellente isolation thermique et acoustique, une forte absorption des liquides et une diffusion complexe de la lumière. Ce spectre de propriétés fait de la mousse de polyuréthane un des matériaux couramment utilisés dans de nombreuses applications phoniques, thermiques et de confort. Pour contrôler la vibration transmise aux occupants des sièges, plusieurs dispositifs automatiques de régulation et de contrôle sont actuellement en cours de développement tels que les amortisseurs actifs et semi-actifs. La performance de ces derniers dépend bien évidemment de la prédiction des comportements de tous les composants du siège et en particulier la mousse. D’une façon générale, il est indispensable de modéliser le comportement mécanique complexe de la mousse de polyuréthane et d’identifier ses propriétés quasi-statique et dynamiques afin d’optimiser la conception des systèmes incluant la mousse en particulier l’optimisation de l’aspect confort. Dans cette optique, l’objectif principal de cette thèse consiste à implémenter des modèles mécaniques de la mousse de polyuréthane fiables et capables de prévoir sa réponse sous différentes conditions d’essais. Dans la littérature, on retrouve les divers modèles développés tels que les modèles de mémoire entier et fractionnaire. L’inconvénient majeur de ces modèles est lié à la dépendance de leurs paramètres vis-à-vis des conditions d’essais, chose qui affecte le caractère général de leur représentativité des comportements quasi-statique et dynamique de la mousse polyuréthane. Pour pallier à cet inconvénient, nous avons développé des modèles qui, grâce à des choix judicieux de méthodes d’identification, assurent une représentativité plus générale des comportements quasi-statique et dynamique de la mousse polyuréthane. En effet, nous avons démontré qu’on peut exprimer les paramètres dimensionnels des modèles développés par le produit de deux parties indépendantes ; une regroupant les conditions d’essais et une autre définissant les paramètres adimensionnels et invariants qui caractérisent le matériau. Ces résultats ont été obtenus à partir de plusieurs études expérimentales qui ont permis l’appréhension du comportement quasi-statique (à travers des essais de compression unidirectionnelle) et dynamique (à travers des tests en vibration entretenue). La mousse, sous des grandes déformations, présente à la fois un comportement élastique non linéaire et un comportement viscoélastique. En outre, une discrimination entre les modèles développés particulièrement en quasi-statique a été effectuée. Les avantages et les limites de chacun y ont été discutés. / Polyurethane foam is a cellular material characterized by an interesting mechanical spectrum of properties: low density, capacity to absorb the deformation energy and low stiffness. It presents also several other properties, such as excellent thermal and acoustic insulation, high absorption of fluids and a complex scattering of light. This spectrum of properties makes polyurethane foam commonly used in many thermal, acoustic and comfort applications. To control the vibration transmitted to the seat occupants, several automatic devices for regulation and control are currently outstanding developments like active and semi-active dampers. The performance of these devices depends, of course, on the prediction of the behaviour of all the seat components and especially foam. Generally, it is essential to model the complex mechanical behaviour of polyurethane foam and identify its quasi-staticand dynamicproperties in order to optimize the design of systems with foam particularly the optimization of the comfort aspect. In this mind, the main goal of this thesis is to implement mechanical models of polyurethane foam reliable and able to provide its response under different test conditions. Several models has been developed in literature such as memory fractional and integer models. The main disadvantage of these models is the dependence of their parameters against the test conditions. It affects the general character of their representativeness to the quasi-static and dynamic behaviours of polyurethane foam. To solve this problem, we developed models with specific identification methods to ensure broader representation of the quasi-static and dynamic behaviour of polyurethane foam. Indeed, we have demonstrated that we can express the dimensional parameters of the developed models by the product of two independent parts; the first contain only the test conditions and the second define the dimensionless and invariant parameters that characterize the foam material. The developed models have been establish after several experimental studies allowing the apprehension of the quasi-static behaviour (through unidirectional compression tests) and the dynamic behaviour (through harmonic vibration tests). The polyurethane foam, under large deformations, exhibits a non linear elastic behaviour and viscoelastic behaviour. In addition, discrimination between the models developed especially in quasi-static case has been conducted. The advantages and limitations of each model have been discussed.

La mousse flexible de polyuréthane est couramment utilisée dans nombreuses applications telles que acoustiques, thermiques et de bâtiment grâce à sa faible densité et à son pouvoir d’isolation thermique et acoustique. Elle est également utilisée dans les applications de confort pour les sièges tels que véhicule, train, avion etc. grâce sa faible raideur et à son pouvoir à absorber l’énergie de déformation. Pour optimiser le confort des systèmes d’assise, il est nécessaire de modéliser le siège et en particulier la partie flexible, c’est-à-dire la mousse de polyuréthane. Les objectifs principaux de cette thèse consistent à identifier puis à modéliser le comportement quasi-statique de la mousse de polyuréthane sous différentes conditions d’essais sous grandes déformations. Des essais de compression/décompressions unidirectionnels monocycle et multicycle à différentes vitesses de déformations ont été réalisés sur trois types de mousse de polyuréthane, afin de comprendre le comportement du matériau. Ces essais ont permis de déduire que les mousses de polyuréthanes sous grandes déformations présentent à la fois un comportement hyperélastique et un comportement viscoélastique. Ils ont également montrés que les mousses de polyuréthanes présentent un phénomène d’assouplissement appelé ‘effet de Mullins’ lors que les essais de compression/décompressions multicycle, c’est-à-dire que les contraintes dans 1er cycle sont moins faibles que les contraintes dans les cycles suivants pour une même déformation. Sur la base des résultats d’expérimentaux et afin de modéliser le comportement quasi-statique de la mousse de polyuréthanne, nous avons développé trois modèles visco-hyperélastiques qui se composent de deux éléments à savoir la partie modèles énergétiques hyperélastiques, utilisés généralement pour des matériaux à comportement caoutchoutique, et la partie modèle à mémoire entier qui tient compte de l’historique et permettant de décrire le comportement viscoélastique. Les paramètres des modèles ont été identifiés en utilisant la méthode d’identification et la méthode d’optimisation appropriée. Les résultats des modélisations du comportement mécanique de la mousse sur les essais monocycles et multicycles ont été comparés aux résultats expérimentaux, monteront à la fois une très bonne capacité à simuler le 1er cycle de charge/décharge, ainsi que les cycles suivant. Nos modèles ont prouvé leur capacité à modéliser l’effet de Mullins sur les mousses de polyuréthane souple. Ces modèles ont été validés sur les trois types de mousse et pour trois vitesses de sollicitation, permettent de conclure leurs efficacités et de leurs représentativités. / Flexible polyurethane foam is widely used in numerous applications such as acoustic, thermal and building due to its low density and its ability to absorb thermal and acoustic energy. It is also used for the comfort of the seats such as the vehicle, train, plane due to its low stiffness and its ability to absorb deformation energy. In order to optimize the comfort of the car seat, it is necessary to model the behavior of seat system, particularly the flexible component - polyurethane foam. The main objective of this study is to identify and model the quasi-static behavior of polyurethane foam under different test conditions in large deformations. Compression / decompression uniaxial unicycle and multicycle tests were carried out on three types of polyurethane foam at different strain rates, which allows us to understand the behavior of the material. The results of the tests indicate that the polyurethane foams exhibit a hyperelastic behavior and a viscoelastic behavior under large deformations. They also showed that the polyurethane foams have a stress softening phenomenon which is called 'Mullins effect' during the compression / decompression multicycle tests. In other words, the stress in first cycle is lower than the stresses in the subsequent cycles in the same deformation. ‘Mullins effect’ for the polyurethane foam is also an important study in this dissertation. Based on the experimental results and the goal of modeling quasi-static behavior of the polyurethane foam, three visco-hyperelastic models were developed. These models consist in two elements: hyperelastic models, which is normally used for description the behavior of rubber materials, and entire memory model which takes into account the history and describing the viscoelastic behavior. Model parameters were identified using appropriate identification and optimization methods. The results of modeling the mechanical behavior of the foam on the unicycle and multicycle tests were compared with experimental results. The models showed a very good competence to simulate the first cycle and the following cycles during the charge / discharge tests. Our models have proven its ability to model Mullins effect on flexible polyurethane foams. These models have been validated on three types of foam in order to present a comparative study of their effectiveness and their representativeness.

In dieser Arbeit wurde die Ionisation einiger zweiatomiger Moleküle (H2, N2 und O2) in intensiven Laserfeldern untersucht. Hierbei wurden verschiedene Modelle zur Beschreibung der Tunnelionisation sowohl untereinander als auch mit der Lösung der zeitabhängigen Schrödingergleichung (TDSE) verglichen. Die kernabstandsabhängige Ionisationswahrscheinlichkeit wurde für verschiedene Intensitäten betrachtet und die Gültigkeit modifizierter atomarer bzw. Molekularer Modelle zur Beschreibung der Tunnelionisation analysiert. Es wurde herausgefunden, dass Modelle, die auf der quasistatischen Näherung beruhen (wo die Ionisation unabhängig von der Frequenz des Laserfeldes ist), nur in einem kleinem Frequenz- und Intensitätsbereich hinreichend genaue Ergebnisse liefern, dem Tunnelregime. Modelle mit einem frequenzabhängigen Faktor stimmen hingegen sowohl im Tunnel- als auch im Mehrphotonenregime mit den genaueren TDSE Ergebnissen überein. Weiterhin wird auch die Abweichung zur Franck-Condon Näherung verdeutlicht. Es wurde ein kleiner Einfluss auf die Revival-Zeit des im Wasserstoffmolekül-Ion gestarteten Wellenpakets gefunden. Die Berücksichtigung von Bond-Softening führt weiterhin zu einer Verringerung der Revival-Zeit mit steigender Spitzenintensität des Lasers. Außerdem wird die Anisotropie der Ionisation von H2 als Funktion der Laserintensität in linear und zirkular polarisiertem Licht mit dem molekularen Tunnelmodell MO-ADK untersucht. Gute Übereinstimmung mit den experimentellen Beobachtungen wurde gefunden, insbesondere wenn der Effekt des Fokusvolumens des Laserfeldes berücksichtigt wird. Die Anwendbarkeit des Zwei-Zentren-Modells auf größere Moleküle, N2 und O2, wird ebenfalls getestet. Es wird beobachtet, dass dies für N2 (symmetrisches HOMO) funktioniert, für O2 (asymmetrisches HOMO) jedoch nicht. / The ionization of some diatomic molecules, H2, N2, and O2, exposed to intense laser fields has been studied by comparing various molecular tunneling–ionization models with each other and with the numerical solution of the time-dependent Schrödinger equation (TDSE). The internuclear-distance dependent ionization yields over a wide range of laser peak intensities are investigated and the validity of the modified atomic and molecular tunneling models is examined. It is found that those models that depend on the quasi-static approximation, where ionization is independent on the oscillation frequency of the applied laser field, are useful for laser-induced ionization processes in only a very small region of the frequency and intensity domain of laser fields, i.e. in the tunneling regime. The models that include a frequency dependent factor are in a good agreement with the accurate TDSE calculations in both the multiphoton and the tunneling ionization regimes. Furthermore, the deviation from Franck-Condon-like distribution is also clarified. A small effect on the revival time of the vibrational wavepacket of hydrogen molecular ion, due to this deviation, has been found. Consideration of the bond-softening effect leads to a decrease in the revival time with increasing laser-peak intensity. The anisotropy of H2 as a function of laser intensity in linear and circular polarized fields using molecular tunneling model (MO-ADK) are also studied and a good agreement with the experimental observations, especially if the focal volume of the laser field is considered, has been obtained. The applicability of the two-center model for larger molecules, N2 and O2, is tested. It is found that it works with N2 (symmetric HOMO) but fails in O2 (ansymmetric HOMO).

The presented work is concerned with the development and application of an ultrasonic detection scheme suitable for the monitoring of muscle dynamics with high temporal - down to 5 µs - and spatial resolution - down to 0.78 µm. A differential detection scheme has been developed to monitor the variations of the velocity of longitudinal polarized ultrasound waves travelling in contracting and relaxing muscle, compensating for variations of the path length by referencing to a frame. The observed time dependent variations of the time-of-flight of the ultrasonic waves caused by variations in the muscle and in addition by minor deformations of the enclosure are detected each separately and synchronously and are evaluated differentially. Beside of the detected increase of the speed of sound observed for contracted muscle with respect to the relaxed state of about 0.6%, the recovery time from maximum isometric contraction is quantified and relaxation processes are observed for the recovery phase following the isometric contraction. The developed ultrasonic calliper was employed to monitor both, the brain controlled and externally excited muscle dynamics with sampling intervals down to 10 ms synchronously with signals relating to the excitation. Monitored are the activation, hold, and relaxation phase for maximum voluntary isometric contraction of the gastrocnemius muscle. A so far not reported post tetanus overshoot and subsequent exponential recovery are observed. Both are attributed to the muscle as suggested by combined monitoring with EMG and are modelled with a lumped mechanical circuit containing an idealized bidirectional linear motor unit, ratchet, damper, and springs. Both, the rapid contraction and relaxation phases require a high order filter or alternatively a kernel filter, attributed to the nerve system as suggested by external electric stimulation. The respective response function is modelled by an electrical lumped circuit. Together with a reaction time and occasionally observed droops in the hold phase, both adjusted empirically, the monitored response is represented in close approximation by the combined electrical and mechanical lumped circuits. The respectively determined model parameters provide a refined evaluation scheme for the performance of monitored athletes. Valuable parameters relate to the latent period, the muscle response time, the activation and deactivation dynamics, a possible droop and other instabilities of the hold phase, and parameters characterizing the relaxation phase including the observed post tetanus overshoot and subsequent contraction. Monitored and modelled are also the different processes involved in active muscle dynamics including isotonic, isometric, and eccentric contraction or stretching. The developed technology provides time sequential observation of these processes and registration of their path in the extension and force parameter space. Under suitable conditions the closed-loop cycles of mind controlled human muscle movements proceed along characteristic lines coinciding with well identifiable elementary processes. The presentation of the monitored processes in the extension and force parameter space allows the determination of the mechanical energy expenditure for the observed different muscle actions. An elementary macroscopic mechanical model has been developed, suitable to express the basic features of the monitored muscle dynamics.

Les travaux de thèse visent à améliorer les connaissances sur le comportement parasismique des bâtiments à ossature en bois. Le comportement de ces bâtiments sous sollicitations sismiques est lié à celui de ses assemblages par connecteurs métalliques (pointes, vis, équerres, etc.). La modélisation numérique d'une telle structure s'appuie sur une démarche multi-échelles, afin de représenter les comportements locaux à l'échelle de l'ouvrage. Trois échelles sont définies. Échelle 1 : assemblages, échelle 2 : éléments de structure (mur, plancher, toiture), échelle 3 : bâtiment. A chaque échelle, une loi de comportement dédiée (hystérétique avec endommagement) permet la modélisation. Les calages ou validations s'appuient sur des campagnes d'essais expérimentaux. Diverses configurations de spécimen et divers chargements sont testés afin de construire une vaste base de données de résultats. Les essais sur les assemblages par connecteurs métalliques ont permis le calage du modèle numérique à l'échelle 1. Un modèle éléments finis (EF) détaillé de mur est validé expérimentalement en quasi-statique et en dynamique. Un modèle EF simplifié de mur (macro) permet de passer à l'échelle du bâti. Cet élément macro, calibré sur le modèle détaillé, permet de reproduire avec une précision satisfaisante le comportement dynamique d'un mur. L'assemblage d'éléments de murs (pleins ou avec ouverture) permet de tendre vers la modélisation tridimensionnelle d'une structure. Ce modèle numérique de structure permettra d'étudier, localement et globalement, le comportement parasismique des constructions à ossature bois afin de proposer des dispositifs constructifs et un dimensionnement adaptés à ces ouvrages en zone sismique. / This research aims at a better understanding of the vulnerability of timber-frame buildings against earthquakes. The behavior of these structures under seismic loading relies on their joints with metal fasteners (nails, screws, 3 dimensionnal connections, etc.). The numerical modeling of such a structure is based on a multi-scale approach, which allows to take into account the local behaviors at the structural scale. Three scales are defined: Scale 1: joints, scale 2: structural elements (shear walls, floors, roofs), scale 3: buildings. At each scale, a behavior law (hysteretic with damage) is used. The calibrations or validations are based on experimental tests. Tests on joints with metal fasteners are used to calibrate the numerical model at scale 1. A detailed finite elements (FE) model of shear wall is developped and its predictions are confronted to quasi-static and dynamic experimental results for validation. A simplified FE model of shear wall (macro element) is used to generate a numerical model at the building scale. This macro element, calibrated on the detailed FE model, accurately reproduces the dynamic behavior of a shear wall despite its simplicity. The numerical model of timber-frame buildings will be used to study, locally and globally, their behavior against earthquake in order to propose construction details and design adapted to these structures in seismic areas.

Afin d'estimer les forces statiques équivalentes du vent, qui produisent les réponses quasi-statiques et dynamiques extrêmes dans les structures soumises au champ de pression instationnaire induit par les effets du vent, une nouvelle méthode probabiliste est proposée. Cette méthode permet de calculer les forces statiques équivalentes du vent pour les structures avec des écoulements aérodynamiques complexes telles que les toitures de stade, pour lesquelles le champ de pression n'est pas gaussien et pour lesquelles la réponse dynamique de la structure ne peut être simplement décrite en utilisant uniquement les premiers modes élastiques (mais nécessitent une bonne représentation des réponses quasi-statiques). Généralement, les mesures en soufflerie du champ de pression instationnaire appliqué à une structure dont la géométrie est complexe ne suffisent pas pour construire une estimation statistiquement convergée des valeurs extrêmes des réponses dynamiques de la structure. Une telle convergence est nécessaire pour l'estimation des forces statiques équivalentes afin de reproduire les réponses dynamiques extrêmes induites par les effets du vent en tenant compte de la non-gaussianité du champ de pression aléatoire instationnaire. Dans ce travail, (1) un générateur de réalisation du champ de pression instationnaire non gaussien est construit en utilisant les réalisations qui sont mesurées dans la soufflerie à couche limite turbulente; ce générateur basé sur une représentation en chaos polynomiaux permet de construire un grand nombre de réalisations indépendantes afin d'obtenir la convergence des statistiques des valeurs extrêmes des réponses dynamiques, (2) un modèle d'ordre réduit avec des termes d'accélération quasi-statique est construit et permet d'accélérer la convergence des réponses dynamiques de la structure en n'utilisant qu'un petit nombre de modes élastiques, (3) une nouvelle méthode probabiliste est proposée pour estimer les forces statiques équivalentes induites par les effets du vent sur des structures complexes décrites par des modèles éléments finis, en préservant le caractère non gaussien et sans introduire le concept d'enveloppes des réponses. L'approche proposée est validée expérimentalement avec une application relativement simple et elle est ensuite appliquée à une structure de toiture de stade pour laquelle des mesures expérimentales de pressions instationnaires ont été effectuées dans la soufflerie à couche limite turbulente / In order to estimate the equivalent static wind loads, which produce the extreme quasi-static and dynamical responses of structures submitted to random unsteady pressure field induced by the wind effects, a new probabilistic method is proposed. This method allows for computing the equivalent static wind loads for structures with complex aerodynamic flows such as stadium roofs, for which the pressure field is non-Gaussian, and for which the dynamical response of the structure cannot simply be described by using only the first elastic modes (but require a good representation of the quasi-static responses). Usually, the wind tunnel measurements of the unsteady pressure field applied to a structure with complex geometry are not sufficient for constructing a statistically converged estimation of the extreme values of the dynamical responses. Such a convergence is necessary for the estimation of the equivalent static loads in order to reproduce the extreme dynamical responses induced by the wind effects taking into account the non-Gaussianity of the random unsteady pressure field. In this work, (1) a generator of realizations of the non-Gaussian unsteady pressure field is constructed by using the realizations that are measured in the boundary layer wind tunnel; this generator based on a polynomial chaos representation allows for generating a large number of independent realizations in order to obtain the convergence of the extreme value statistics of the dynamical responses, (2) a reduced-order model with quasi-static acceleration terms is constructed, which allows for accelerating the convergence of the structural dynamical responses by using only a small number of elastic modes of the structure, (3) a novel probabilistic method is proposed for estimating the equivalent static wind loads induced by the wind effects on complex structures that are described by finite element models, preserving the non-Gaussian property and without introducing the concept of responses envelopes. The proposed approach is experimentally validated with a relatively simple application and is then applied to a stadium roof structure for which experimental measurements of unsteady pressures have been performed in boundary layer wind tunnel

Cette thèse vise à étudier le comportement de revêtement articulé du tunnel en développant une nouvelle approche numérique à la Méthode de Réaction Hyperstatique (HRM) et la production des modèles numériques en deux dimensions et trois dimensions à l'aide de la méthode des différences finies (FDM). L'étude a été traitée d'abord sous charges statiques, puis effectuée sous charges dynamiques. Tout d'abord, une étude bibliographique a été effectuée. Une nouvelle approche numérique appliquée à la méthode HRM a ensuite été développée. En même temps, un modèle numérique en deux dimensions est programmé sur les conditions de charge statique dans le but d'évaluer l'influence des joints, en termes de la distribution et des caractéristiques des joints, sur le comportement du revêtement articulé de tunnel. Après cela, des modèles complets en trois dimensions d'un seul tunnel, de deux tunnels horizontaux et de deux tunnels empilés, dans lesquels le système des joints est simulé, ont été développés. Ces modèles en trois dimensions permettent d'étudier le comportement non seulement du revêtement du tunnel, mais encore le déplacement du sol entourant le tunnel lors de l’excavation. Un modèle numérique en trois dimensions simplifié a ensuite été réalisé afin de valider la nouvelle approche numérique appliquée à la méthode HRM.Dans la dernière partie de ce mémoire, la performance du revêtement articulé du tunnel sous chargements dynamiques est prise en compte par l’analyse quasi-statique et dynamique complète en utilisant le modèle numérique en deux dimensions (FDM). Un modèle HRM a également été développé prenant en compte des charges quasi-statiques. Les différences de comportement de tunnel sous chargements statiques et sismiques sont mises en évidence et expliquées. / This PhD thesis has the aim to study the behaviour of segmental tunnel lining by developing a new numerical approach to the Hyperstatic Reaction Method (HRM) and producing two-dimensional (2D) and three-dimensional (3D) numerical models using the finite difference method (FDM). The study first deals with under static loads, and then performs under dynamic loads. Firstly, a literature review has been conducted. A new numerical approach applied to the HRM has then been developed. At the same time, a 2D numerical model is programmed regarding static loading conditions in order to evaluate the influence of the segmental joints, in terms of both joint distribution and joint stiffness characteristics, on the tunnel lining behaviour. After that, full 3D models of a single tunnel, twin horizontal tunnels and twin tunnels stacked over each other, excavated in close proximity in which the joint pattern is simulated, have been developed. These 3D models allow one to investigate the behaviour of not only the tunnel lining but also the displacement of the ground surrounding the tunnel during the tunnel excavation. A simplified 3D numerical model has then been produced in order to validate the new numerical approach applied to the HRM. In the last part of the manuscript, the performance of the segmental tunnel lining exposed to dynamic loading is taken into consideration through quasi-static and full dynamic analyses using 2D numerical models (FDM). A new HRM model has also been developed considering quasi-static loads. The differences of the tunnel behaviour under static and seismic loadings are highlighted.

Since the mid 1990’s, the wireless communications industry has witnessed explosive growth. The worldwide cellular and personal communication subscriber base surpassed 600 million users by late 2001, and the number of individual subscribers surpassed 2 billion at the end of 2006 [1, 2]. In order to attract and accommodate these subscribers, modern communication systems, like the Third Generation (3G) and Fourth Generation (4G) cellular networks, will have to provide attractive new features such as increased data throughput rates, greater system capacity, and better speech quality. These modern communication systems promise to have advantages such as wireless access in ways that have never been possible before, providing, amongst others services such as live television (TV) broadcasting to Mobile Stations (MS)s, multi-megabit Internet access, communication using Voice over Internet Protocol (VoIP), unparalleled network capacity, seamless accessibility and many more. With specific, but not exclusive reference to the cellular environment, there are numerous ways to increase the data throughput rate and system capacity. From an economical perspective, it would be more efficient to add equipment to the Base Station (BS) rather than the MSs. To achieve these improvements the motivation to utilise transmit diversity’s capabilities have been identified as a key research issue in this study. Alamouti [3] proposed a transmit diversity technique using two transmit antennas and one receive antenna, providing the same diversity order than using one transmit antenna and two receive antennas. Since Alamouti’s publication in 1998, many papers in the field of Space-Time (ST) coding have been published. Current research in the field of ST coding consists of finding methods to extend the number of transmit antennas to more than four, while still achieving full rate, as well as full diversity which is the main motivation for this study. This study proposes a novel idea of breaching the limitations with ST coding theory by combining ST coding with Spread Spectrum (SS) modulation techniques in order to extend the number of transmit antennas to more than four and still achieve full rate as well as full diversity. An advantage of the proposed scheme, called Direct Sequence Space-Time Spreading (DSSTS) has over current Space-Time Spreading (STS) techniques is that it uses 50% less spreading codes. A performance evaluation platform for the DSSTS scheme was developed to simulate the performance of the scheme in a realistic mobile communication environment. A mobile communication channel that has the ability to simulate time-varying multipath fading was developed and used to evaluate the performance of the DSSTS scheme. From the simulation results obtained, it is evident that Walsh sequences that exhibit particularly good cross-correlation characteristics, cannot overcome the effect of the antenna self-noise in order to exploit the diversity gain by adding extra antennas, i.e. diversity extension. The research also showed that an optimal trade-off exists between antenna diversity and antenna created self-noise. Performance results of the DSSTS scheme in slow and fast fading channels for a different number of transmit antennas are also presented in this study. With the capacity analysis of the DSSTS scheme, it was shown that the addition of extra transmit antennas to the system indeed increased the system capacity. A further addition to this study is the investigation into the assumption that the channel should be quasi-static over the frame length of the ST code. A Space Sequence Transmit Diversity (SSTD) technique is consequently proposed that allows the transmission of the Alamouti symbols during one time interval instead of two. This relieves the ST code from the assumption that the channel should be quasi-static, allowing it to be used in a more realistic multi-user environment. A performance evaluation platform for the SSTD scheme was developed and used to obtain simulation results in a multipath fading channel. It was also shown that the proposed SSTD scheme is successful in combating the effects of multipath fading for small Code Division Multiple Access (CDMA) user loads. However, as a rule of thumb, the square root of the spreading sequence length divided by two depicts the user load at which the SSTD scheme was not capable of overcoming the combined effects of Multi-User Interference (MUI) and multipath fading. / Dissertation (MEng)--University of Pretoria, 2008. / Electrical, Electronic and Computer Engineering / unrestricted