Orientador: Prof. Dr. Fernando Luis da Silva Semião / Dissertação (mestrado) - Universidade Federal do ABC, Programa de Pós-Graduação em Física, 2017.

abstract: Two major challenges in the transformer-less, single-phase PV string inverters are common mode leakage currents and double-line-frequency power decoupling. In the proposed doubly-grounded inverter topology with innovative active-power-decoupling approach, both of these issues are simultaneously addressed. The topology allows the PV negative terminal to be directly connected to the neutral, thereby eliminating the common-mode ground-currents. The decoupling capacitance requirement is minimized by a dynamically-variable dc-link with large voltage swing, allowing an all-film-capacitor implementation. Furthermore, the use of wide-bandgap devices enables the converter operation at higher switching frequency, resulting in smaller magnetic components. The operating principles, design and optimization, and control methods are explained in detail, and compared with other transformer-less, active-decoupling topologies. A 3 kVA, 100 kHz single-phase hardware prototype at 400 V dc nominal input and 240 V ac output has been developed using SiC MOSFETs with only 45 μF/1100 V dc-link capacitance. The proposed doubly-grounded topology is then extended for split-phase PV inverter application which results in significant reduction in both the peak and RMS values of the boost stage inductor current and allows for easy design of zero voltage transition. A topological enhancement involving T-type dc-ac stage is also developed which takes advantage of the three-level switching states with reduced voltage stress on the main switches, lower switching loss and almost halved inductor current ripple. In addition, this thesis also proposed two new schemes to improve the efficiency of conventional H-bridge inverter topology. The first scheme is to add an auxiliary zero-voltage-transition (ZVT) circuit to realize zero-voltage-switching (ZVS) for all the main switches and inherent zero-current-switching (ZCS) for the auxiliary switches. The advantages include the provision to implement zero state modulation schemes to decrease the inductor current THD, naturally adaptive auxiliary inductor current and elimination of need for large balancing capacitors. The second proposed scheme improves the system efficiency while still meeting a given THD requirement by implementing variable instantaneous switching frequency within a line frequency cycle. This scheme aims at minimizing the combined switching loss and inductor core loss by including different characteristics of the losses relative to the instantaneous switching frequency in the optimization process. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2017

abstract: A new photovoltaic (PV) array power converter circuit is presented. The salient features of this inverter are: transformerless topology, grounded PV array, and only film capacitors. The motivations are to reduce cost, eliminate leakage ground currents, and improve reliability. The use of Silicon Carbide (SiC) transistors is the key enabling technology for this particular circuit to attain good efficiency. Traditionally, grid connected PV inverters required a transformer for isolation and safety. The disadvantage of high frequency transformer based inverters is complexity and cost. Transformerless inverters have become more popular recently, although they can be challenging to implement because of possible high frequency currents through the PV array's stay capacitance to earth ground. Conventional PV inverters also typically utilize electrolytic capacitors for bulk power buffering. However such capacitors can be prone to decreased reliability. The solution proposed here to solve these problems is a bi directional buck boost converter combined with half bridge inverters. This configuration enables grounding of the array's negative terminal and passive power decoupling with only film capacitors. Several aspects of the proposed converter are discussed. First a literature review is presented on the issues to be addressed. The proposed circuit is then presented and examined in detail. This includes theory of operation, component selection, and control systems. An efficiency analysis is also conducted. Simulation results are then presented that show correct functionality. A hardware prototype is built and experiment results also prove the concept. Finally some further developments are mentioned. As a summary of the research a new topology and control technique were developed. The resultant circuit is a high performance transformerless PV inverter with upwards of 97% efficiency. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2014

Nous nous intéressons dans ce mémoire à des méthodes de parallélisation par découplage du système dynamique. Plusieurs applications numériques de nos jours conduisent à des systèmes dynamiques de grande taille et nécessitent des méthodes de parallélisation en conséquence pour pouvoir être résolues sur les machines de calcul à plusieurs processeurs. Notre but est de trouver une méthode numérique à la fois consistante et stable pour réduire le temps de la résolution numérique. La première approche consiste à découpler le système dynamique en sous-systèmes contenant des sous-ensembles de variables indépendants et à remplacer les termes de couplage par l’extrapolation polynomiale. Une telle méthode a été introduite sous le nom de schéma C (p, q, j), nous améliorons ce schéma en introduisant la possibilité à utiliser des pas de temps adaptatifs. Cependant, notre étude montre que cette méthode de découplage ne peut satisfaire les propriétés numériques que sous des conditions très strictes et ne peut donc pas s’appliquer aux problèmes raides présentant des couplages forts entre les sous-systèmes. Afin de pouvoir répondre à cette problématique de découplage des systèmes fortement couplés, on introduit le deuxième axe de recherche, dont l’outil principal est la réduction d’ordre du modèle. L’idée est de remplacer le couplage entre les sous-ensembles de variables du système par leurs représentations sous forme réduite. Ces sous-systèmes peuvent être distribués sur une architecture de calcul parallèle. Notre analyse du schéma de découplage résultant nous conduit à définir un critère mathématique pour la mise à jour des bases réduites entre les sous-systèmes. La méthode de réduction d’ordre du modèle utilisée est fondée sur la décomposition orthogonale aux valeurs propres (POD). Cependant, ne disposant pas à priori des données requises pour la construction de la base réduite, nous proposons alors un algorithme de construction incrémentale de la base réduite permettant de représenter le maximum des dynamiques des solutions présentes dans l’intervalle de simulation. Nous avons appliqué la méthode proposée sur les différents systèmes dynamiques tels que l’exemple provenant d’une EDP et celui provenant de l’équation de Navier Stokes. La méthode proposée montre l’avantage de l’utilisation de l’algorithme de découplage basé sur la réduction d’ordre. Les solutions numériques sont obtenues avec une bonne précision comparées à celle obtenue par une méthode de résolution classique tout en restant très performante selon le nombre de sous-systèmes définis. / In this thesis, we are interested in parallelization algorithm for solving dynamical systems. Many industrial applications nowadays lead to large systems of huge number of variables. A such dynamical system requires parallel method in order to be solved on parallel computers. Our goal is to find a robust numerical method satisfying stability and consistency properties and suitable to be implemented in parallel machines. The first method developed in this thesis consists in decoupling dynamical system into independent subsystems and using polynomial extrapolation for coupled terms between subsystems. Such a method is called C(p; q; j).We have extended this numerical scheme to adaptive time steps. However, this method admits poor numerical properties and therefore cannot be applied in solving stiff systems with strong coupling terms.When dealing with systems whose variables are strongly coupled, contrary to the technique of using extrapolation for coupled terms, one may suggest to use reduced order models to replace those terms and solve separately each independent subsystems. Thus, we introduced the second approach consisting in using order reduction technique in decoupling dynamical systems. The order reduction method uses the Proper Orthogonal Decomposition. Therefore, when constructing reduced order models, we do not have all the solutions required for the POD basis, then we developed a technique of updating the POD during the simulation process. This method is applied successfully to solve different examples of dynamical systems : one example of stiff ODE provided from PDE and the other was the ODE system provided from the Nervier-Stokes equations. As a result, we have proposed a robust method of decoupling dynamical system based on reduced order technique. We have obtained good approximations to the reference solution with appropriated precision. Moreover, we obtained a great performance when solving the problem on parallel computers.

O viés da tarifa fixa ocorre quando um consumidor contrata um determinado nível de serviço e, em média, consome menos do que ele/ela tem direito a consumir. Tal comportamento é documentado em variados setores, em especial no de telecomunicações. Nesta dissertação, espera-se que o tipo de tarifa que uma pessoa tem seja a causa de determinados comportamentos. Assim, este estudo se baseia nas teorias da contabilidade mental e da dissociação da transação, para medir o impacto de tipos diferentes de tarifas no comportamento de consumo de telefonia móvel. Em uma sequência de três experimentos, foi descoberto que usuários de tarifa fixa são mais dispostos a abrir mão do consumo e se sentem menos frustrados e menos arrependidos do que usuários de tarifa por uso. Este comportamento é condizente com os efeitos dos custos irrecuperáveis que, conforme hipotetizado, também são observados em serviços por assinatura, e não apenas transacionais. Adicionalmente, foi observado que na medida em que aumenta a quantidade de pessoas consumindo o mesmo plano, a disposição a renunciar o consumo permanece a mesma. Por fim, foi observado que o viés da tarifa fixa não compromete comportamentos futuros, como a intenção de migrar ou permanecer com o mesmo plano. Estes resultados apoiam tanto implicações gerenciais quanto teóricas, e oferecem novos caminhos de estudos em mercados menos típicos, porém promissores, de serviços por assinatura. / The flat-rate bias takes place whenever a consumer contracts a given service level and, in average, consumes only a fraction of what s/he is entitled. Such behavior is documented in a number of different sectors, specially the telecommunications sector. In this thesis, the type of tariff an individual has is expected to be the cause of particular behaviors. Therefore, this study draws on mental accounting and transaction decoupling theories, to measure the impact of different types of tariffs on mobile phone consumer behavior. In a series of three experiments, it was found that flat-rate consumers are more willing to forgo consumption and they feel less frustrated and less regretful than users of a pay-per-use plan. This behavior is consistent with the sunk cost effects which, as hypothesed, can also be seen in contract services, and not only in transactional ones. Additionally, it was noted that as the amount of people consuming the same plan increases, the willingness to forgo consumption remains unaltered. Finally, it was observed that the flat-rate bias does not commit future behavior, such as the intention to either change or stay with the same plan. These results support both managerial and theoretical implications, and offer new avenues of studies on the less typical, yet promising, markets of subscription services.

Atualmente se observa o incremento exponencial das exigÃncias pela otimizaÃÃo de desempenho em todos os setores da vida moderna. Esta busca por melhor desempenho tambÃm tem ocorrido em todos os setores da indÃstria, seja ela de atividade fabril ou de energia. Melhor desempenho no setor industrial se reflete em melhorias em todos os nÃveis e isto alcanÃa de sobremaneira os sistemas de controle de processos. Com requisitos de desempenho mais rÃgido, tÃcnicas antigamente estudadas somente em meio acadÃmico tem migrado para a indÃstria abrindo novos horizontes e estabelecendo novos conceitos de uma forma geral. Com os sistemas de controle de processos nÃo à diferente e em particular em seu mais baixo nÃvel de implementaÃÃo (chÃo de fÃbrica) tem se observado uma atualizaÃÃo constante. Esta dissertaÃÃo està inserida neste Ãltimo contexto ou seja no projeto das malhas de controle de um sistema de controle de processos. Desta forma aspectos de sintonia de controladores convencionais do tipo PID (Proporcional, Integral, Derivativo) multivariÃveis (MIMO) e controladores MIMO preditivo baseado em modelos do tipo GPC (Generalized Predictive Control) sÃo estudados visando a implementaÃÃo prÃtica. VÃrios aspectos do sistema de controle tais como a identificaÃÃo do processo atravÃs de tÃcnicas paramÃtricas e nÃo paramÃtricas com diferentes representaÃÃes e aspectos do desacoplamento entre as malhas do processo, por se tratar de sistemas MIMO, sÃo tambÃm abordadas, discutidas e implementadas. Para este propÃsito duas plataformas de desenvolvimento foram desenvolvidas no laboratÃrio do NÃcleo de Engenharia ElÃtrica NEEL do IFPI. A primeira plataforma consiste em sistema MIMO de ventilaÃÃo e aquecimento, a segunda plataforma um sistema de tanques duplo acoplados. Como conclusÃo do estudo dos experimentos prÃticos realizados pode-se estabelecer que o melhor controle para um processo depende da aplicaÃÃo ao qual este à destinado e para o bom funcionamento à necessÃrio o conhecimento teÃrico e prÃtico da dinÃmica do processo. Assim um bom controlador depende tambÃm do emprego de uma tÃcnica de identificaÃÃo adequada. / Nowadays, we observe the exponential increasing of demand for performance optimization in all areas of modern life. This search for better performances is also occurring in all industry sectors, either factory or energy activities. This best performances reflects an increase is all levels and achieve enormously the process control systems. With more strict performance requirement, technique formerly studied only in academic circle has been migrate for industry area, opening new horizons and establishing new concepts in general. It is not different with the process control systems, even in particular at the lowest level of implementation (factoryâs floor) has been noted a constantly update. This paper is inserted into this last concept, in other words, into the project of control mesh of process control systems. Thus the aspect of syntony of conventional controllers like PID (Proportional, Integral, Derivative) multivariable (MIMO) and MIMO predictive controllers based on models like GPC (Generalized Predictive Control) type are studied aiming to the practical implementation. Several system control aspects, such as identification of process through parametric and nonparametric technical with different representations and aspects of decoupling between the process meshes, as it is MIMO system, are also dealt, discussed and implemented. For that purpose, two development platforms were developed at IFPIâs NEE (NÃcleo de Engenharia ElÃtrica) laboratory. The first platform consists of MIMO system of ventilation and heating; the second platform is a system of coupled double tanks. As a conclusion of the study of practical experiments conducted, can be established that the best control for a process depends on the application for which it is destined and for the proper functioning is necessary the theoretical and practical knowledge of the dynamic of the process. Thus a proper controller depends also on the use of a technique for proper identification.

A técnica de dupla ressonância unidimensional para alta resolução em líquidos constitui o objetivo deste trabalho. Para isso construiu-se uma sonda de dupla ressonância, apropriada para experiências heteronucleares do tipo 13C-{1H}. Empregam-se algumas sequências de pulsos para permitir as medidas do EON e do desacoplamento nuclear entre 1H e 13C. Emprega-se o EON com a finalidade de obter um aumento na intensidade do sinal de RMN do 13C; emprega-se o desacoplamento para eliminar a estrutura fina que surge devido ao acoplamento nuclear entre 1H e 13C. Inclui-se uma breve revisão sobre a teoria com a fmalidade de unificar o tratamento desta com as aplicações. As técnicas experimentais são discutidas e alguns espectros são mostrados para ilustrar os principais resultados / The one dimensional double resonance technique for high resolition in liquids is the main purpose of this work. For this we have built a double resonance probe, suitable for heteronuclear 13C-{1H} experiment. Some pulse sequences are employed to allow the NOE and decoupling between 1H e 13C to be measured. The NOE is employed in order to obtain a three-fold enhancement of the 13C NMR signal; decoupling is employed to eliminate the fme structW\'e arising from spin coupling between 1H e 13C. A short review about theory is included to provide unified treatment with applications. The experimental techniques are discussed and some spectra are shown to illustrate the main results.

As mudanças climáticas globais e a expansão da cultura da cana-de-açúcar para regiões com menor suprimento hídrico elevaram a importância de estimativas consistentes das necessidades de água da cultura. Na tentativa de contribuir nesse campo, o presente trabalho buscou analisar o consumo hídrico da cana-de-açúcar em três diferentes escalas espaciais: folha, planta e dossel, além de uma ferramenta de simulação (DSSAT/CANEGRO) quanto ao seu desempenho na estimativa do consumo hídrico da cana-de-açúcar. Foram realizadas medidas de evapotranspiração do dossel pelo método do balanço de energia-razão de Bowen (MRB) e de transpiração das plantas por meio do balanço de calor nos colmos (BC) e das folhas com um analisador de gás à infravermelho (IRGA). Duas áreas experimentais foram utilizadas, sendo a primeira com a variedade CTC 12, na safra 2011/2012, irrigada por gotejamento subsuperficial; a segunda área com a variedade RB867515, irrigada por aspersão por meio de um pivô central, na safra 2012/2013. O acoplamento planta-atmosfera também foi avaliado. As análises do MRB indicaram um fluxo de calor latente médio de 70% da energia disponível no ambiente, com 25% de fluxo de calor sensível e 5% no fluxo de calor no solo. Com a mesma metodologia a evapotranspiração da cultura média (ETc) foi estimada em 3,92 e 3,25 mm dia-1 para as variedades CTC 12 e RB867515, respectivamente, resultando em um coeficiente de cultura (Kc) médio de 1,37, na fase de pleno desenvolvimento vegetativo. No período experimental, o canavial permaneceu mais acoplado à atmosfera (Ω médio=0,37) e a transpiração avaliada pelo BC chegou a 4,7 e 3,62 mm dia-1, respectivamente para CTC 12 e RB867515. A metodologia do IRGA resultou em perda de água ao nível foliar da faixa de até 1 mm h-1. Conforme a demanda atmosférica, o Kc da cultura apresentou-se variável em função da evapotranspiração de referência (ETo). O modelo DSSAT/CANEGRO mostrou-se eficiente na simulação da ETc e transpiração do canavial, com resultados mais satisfatórios (R2=0,59) quando utiliza-se a metodologia FAO 56 nas simulações. / Global climate change and the sugarcane crop expansion to regions with lower water supply became more important the real estimate of crop water requirements. This study aimed to contribute on this regard by analyzing the sugarcane water consumption at three different spatial scales: leaf, plant and canopy. The DSSAT/CANEGRO crop model was also evaluated with respect to sugarcane transpiration and evapotranspiration. Bowen ratio energy balance (BREB) method was performed to measure crop canopy evapotranspiration and the transpiration were measured with sap flow by stem heat balance for plant scale and infra-red gas analyzer (IRGA) for leaf scale. Two experimental sites were used: the first with CTC 12 cultivar and drip irrigation, on 2011/2012 season; in the second experiment the RB867515 variety was grown under sprinkler irrigation by a central pivot, on the 2012/2013 growing season. The plant-atmosphere decoupling factor was also evaluated. BREB method showed latent heat flux representing 70% of the available energy, 25% for sensible heat flux, and 5% for soil heat flux. BREB crop evapotranspiration (ETc) ranged from 3.92 to 3.25 mm day-1 for CTC 12 and RB867515, respectively, resulting in a mean crop coefficient (Kc) of 1.37 at the full vegetative growth stage. In the experimental period, sugarcane crop was coupled to the atmosphere (Ω=0.37) and transpiration assessed by HB reached 4.7 and 3.62 mm day-1 for CTC 12 and RB867515, respectively. IRGA method showed a leaf water loss up to 1 mm h-1. Kc varied with reference evapotranspiration (ETo) The DSSAT/CANEGRO crop model was efficient in the ETc and transpiration simulation, with better results (R2=0.59) when using FAO 56 ETo method in the simulations.

Les réacteurs nucléaires à neutrons rapides sont une solution à long terme pour la production d’énergie car ils valorisent le Plutonium et peuvent utiliser tout le stock de d’Uranium appauvri. Les contraintes actuelles en termes de sûreté requièrent néanmoins des innovations pour favoriser un comportement naturel du cœur lors de transitoires incidentels en augmentant le temps de grâce et la marge à la fusion. Les innovations apportées aux cœurs (hauteur réduite, cœur hétérogène, puissance volumique limitée) induisent un accroissement de la taille des réacteurs. Cette évolution a des conséquences de sûreté essentiellement positives comme la réduction de l’effet de vidange mais d’autres comme la déformation de la nappe de puissance méritent attention. Cette thèse a permis de mieux appréhender ces spécificités au travers du calcul et de l’analyse des harmoniques du flux puis des matrices de fission. Ces études, en soutien au cœur d’ASTRID, ont permis de conforter les options de conception. / The need for energy is a matter of growing concern in the world today, in relation to global climate change. Nuclear energy is of interest because it does not produce greenhouse gases, and it is able to generate a substantial amount of energy at a given time. However, it needs fissile material to operate. Fuel economy is then a sine qua none condition for the development of this energy. Sodium Fast Reactors are a solution for the future of nuclear energy. These reactors are indeed able to use much less Uranium for the same amount of energy released. However, the safety constraints in accordance with todays standards (“forgiving behavior”) require new core designs, which are highly heterogeneous axially and rather flat. Finally, this evolution in reactor design (reduced power density and limited axial height) implies a significant increase in the reactor diameter. It has consequences from both an economic (Pu inventory, vessel size) and operational (power shape stabilization during irradiation) point of view. The understanding of this phenomena is the topic of this PhD.

L'emploi d'un retour visuel dans le but d'effectuer une commande en boucle fermée de robot s'est largement répandu et concerne de nos jours tous les domaines de la robotique. Un tel retour permet d'effectuer une comparaison entre un état désiré et l'état actuel, à l'aide de mesures visuelles. L'objectif principal de cette thèse consiste à concevoir plusieurs types de lois de commande cinématiques par vision stéréo. Ceci concerne aussi l'étude de la stabilité du système en boucle fermée et la convergence des fonctions de tâche. C'est essentiellement le découplage des lois de commandes cinématiques en rotation et en translation qui est recherché ici, selon le nombre d'indices visuels considérés.Les mesures visuelles utilisées dans cette thèse sont les lignes droites 3D. Les intérêts apportés à ce type de mesures visuelles sont la robustesse contre le bruit, et la possibilité de représenter d'autres primitives comme des couples de points ou de plans par la modélisation de Plücker. / Closing the control loop of a manipulator robot with vision feedback is widelyknown. It concerns nowadays all areas of robotics. Such a return can make a comparison between a desired state and current state, using visual measurements. The main objective of this doctoral thesis is to design several types of kinematic control laws for stereo visual servoing. It strongly involves the formalism of the task function which is a well-known and useful mathematical tool to express the visual error as a function of state vectors.We have investigated the decoupling between the rotational and translational velocities control laws together with the epipolar constraint with a stereo visual feedback.That is why, the visual measurements and features used in this thesis are the 3Dstraight lines.The interests of this type of visual features rely on the robustness against the noise, and the possibility to represent straight lines or other features like points or planes pairs by the Plücker coordinates, as a 3D straight line can be represented as well by two points or the intersection of two planes. This makes all the control laws designed in this thesis valid for another visual features like points