A proteção de sistemas de energia elétrica possui papel extremamente importante no aspecto de garantir o fornecimento de energia de maneira segura e confiável. Assim, a ação indevida ou a não atuação deste sistema de proteção pode causar danos materiais ou econômicos tanto para as concessionárias quanto para os consumidores de energia elétrica. Dessa forma, o sistema de proteção deve estar bem ajustado para que possa garantir suas funções, sendo sensível, seletivo, confiável e rápido. Para tanto, uma boa coordenação entre os relés de proteção deve ser estabelecida. No caso de um sistema de transmissão, o qual costuma ser um sistema malhado, a proteção é comumente realizada por relés de distância aliados a relés de sobrecorrente com unidade direcional, sendo que estes funcionam como elemento de retaguarda daqueles. O processo de ajuste desses relés é um trabalho muito difícil e demorado, que pode ainda estar sujeito a erros do engenheiro de proteção responsável pelo estudo. Neste contexto, este trabalho tem como objetivo desenvolver uma metodologia baseada na otimização por enxame de partículas que obtenha automaticamente os ajustes desses relés de forma a garantir a coordenação e seletividade entre eles, tornando assim o processo de ajuste mais rápido e preciso. Dessa forma, essa metodologia pode constituir uma ferramenta de auxílio muito favorável ao engenheiro de proteção. Além disso, como em todo problema de otimização, a função objetivo e as restrições foram definidas de maneira a retratar o problema de coordenação envolvendo tanto os relés de distância quanto os direcionais de sobrecorrente. A metodologia foi aplicada a dois sistemas, um fictício com 16 relés e um sistema de transmissão real com 44 relés, sendo que em ambos os casos ela apresentou resultados bastante satisfatórios proporcionando ajustes bem coordenados. / Electrical power systems protection has a very important role in the aspect of ensuring energy supply with safety and reliability. Thus, improper action or non-actuation of this protection system can cause materials and/or economics damages to electricity utilities as well as ordinary energy consumers. Therefore the protection system must be well adjusted so it can ensure its functions and thus being sensible, selective, reliable and fast. In order to achieve these characteristics, the protective relays must be well coordinated. In the case of meshed transmission systems, the protection is generally performed by distance relays as primary protection associated with directional overcurrent relays as backup protection. The process of adjusting these relays is very tough, slow and it can even be subject to the protection engineer mistakes. In this context, this work aims to develop a particle swarm optimization based methodology that can automatically obtain these relays adjusts so they can ensure relays coordination and selectivity, and then make this setting process faster and more precise. Thus, this methodology may provide a very favorable tool to aid the protection engineer. Moreover, as in any optimization problem, the objective function and the constraints were defined to represent the coordination problem involving both distance and directional overcurrent relays. The methodology was applied to two systems, a fictitious with 16 relays and a real transmission system with 44 relays, and in both cases it showed satisfactory results providing well-coordinated settings.

<p>Предмет истраживања су фракциона и тополошка уопштења једначине<br />телеграфичара као модела електричног вода. Фракционо уопштење се<br />огледа у томе што су класичне конститутивне релације електричних<br />елемената замењене фракционим. Модификација конститутивних<br />релација омогућује да се узму у обзир коначно време релаксације и<br />поларизације диелектрика, као и формирања магнетског поља, односно<br />историја процеса поларизације и магнетизације. Тополошко уопштење<br />огледа се у томе што елементарно коло вода, поред стандардних<br />елемената, има кондензатор и у редној грани којим се моделира ефекат<br />нагомилавања наелектрисања дуж елекричног вода, односно коначно<br />време релаксације.</p> / <p>Predmet istraživanja su frakciona i topološka uopštenja jednačine<br />telegrafičara kao modela električnog voda. Frakciono uopštenje se<br />ogleda u tome što su klasične konstitutivne relacije električnih<br />elemenata zamenjene frakcionim. Modifikacija konstitutivnih<br />relacija omogućuje da se uzmu u obzir konačno vreme relaksacije i<br />polarizacije dielektrika, kao i formiranja magnetskog polja, odnosno<br />istorija procesa polarizacije i magnetizacije. Topološko uopštenje<br />ogleda se u tome što elementarno kolo voda, pored standardnih<br />elemenata, ima kondenzator i u rednoj grani kojim se modelira efekat<br />nagomilavanja naelektrisanja duž elekričnog voda, odnosno konačno<br />vreme relaksacije.</p> / <p>The subject of research is the fractional and topological generalization of the<br />telegraphers&rsquo;s equation in transmission line modeling. The fractional<br />generalization is reflected in the fact that the classical constitutive relations of<br />electrical elements are replaced by the fractional ones. The modification of<br />constitutive relations allows for taking into account the finite relaxation and<br />polarization time of dielectric material, as well as the creation of the magnetic<br />field, it also allows for taking into account the history of polarization and<br />magnetization processes. The topological generalization of the elementary<br />circuit of transmission line, in addition to the standard elements, introduces a<br />capacitor in the series branch, which models the effect of charge<br />accumulation along the transmission line, i.e., the finite relaxation time.</p>

Este trabalho visa apresentar uma análise metodológica para viabilizar a realização da repotenciação, elevando a classe de tensão de linhas de distribuição de 34,5 kV para a classe de subtransmissão de 69 kV. Foi realizado um levantamento de casos de repotenciação de linhas de transmissão, inclusive para a alternativa de elevação da tensão de operação da linha. Foram analisados modelos matemáticos apropriados para a representação da linha frente a transitórios eletromagnéticos como curtos-circuitos e descargas atmosféricas e ainda, um estudo sobre o procedimento a ser realizado dispondo-se de um software baseado na teoria de elementos finitos para a verificação do nível básico de isolamento das estruturas pertencentes à linha. Assim, a linha piloto foi representada no software ATP, utilizando-se a interface ATPDraw, e foi submetida a diversos transitórios eletromagnéticos. A partir dos valores de tensão observados nesses distúrbios, iniciaram-se novas simulações computacionais por meio do software FLUX®, que é baseado na teoria dos elementos finitos, para avaliar o comportamento do campo elétrico nas diferentes estruturas da linha, considerando a cruzeta de madeira e os isoladores cerâmicos. Dessa forma, esse estudo, avalia as simulações computacionais realizadas e aponta quais as estruturas da linha estão aptas a operar na classe de tensão de 69 kV, ou ainda, quais alterações permitirão que estruturas menos seguras possam operar neste novo nível de tensão. Todas as etapas e as conclusões finais sobre o assunto delineado são apresentadas neste documento. / This work presents a methodological analysis to make repowering distribution line possible, raising the voltage level from 34.5 kV to 69 kV (sub transmission level). A literature review of transmission line upgrading was accomplished, including the alternative of raising the line operating voltage. Proper mathematical models were analyzed for the line representation facing electromagnetic transients such as short circuits and lightning. In addition, a software based on finite elements theory for the verification of the basic insulation level of the structures (poles and crossarms) belonging to the line was utilized. Thus, a pilot line was modeled in the ATP software, using the ATPDraw interface, and it was subjected to various electromagnetic transients. From the voltage values observed in these disturbances, new computer simulations were carried out using FLUX®, a finite elements theory based software capable of evaluating the electrical field behavior in different structures of the line, considering wooden crossarms and ceramic insulators. In this context, the present study analyze the computer simulations and points out which line structures are able to operate in the 69 kV voltage level, or even indicates which alterations will enable the less secure structures to operate in this new voltage level. All the stages and final conclusions about the outlined subject are presented in this document.

Communication and reconnaissance systems are requiring increasing flexibility concerning functionality and efficiency for multiband and broadband frequency applications. Circuit-based reconfiguration mechanisms continue to promote radio frequency (RF) application flexibility; however, increasing limitations have resulted in hindering performance. Therefore, the implementation of a "wireless" reconfiguration mechanism provides the required agility and amicability for microwave circuits and antennas without local overhead. The wireless reconfiguration mechanism in this thesis integrates dynamic, fluidic-based material systems to achieve electromagnetic agility and reduce the need for "wired" reconfiguration technologies. The dynamic material system component has become known as electromagnetically functionalized colloidal dispersions (EFCDs). In a microfluidic reconfiguration system, they provide electromagnetic agility by altering the colloidal volume fraction of EFCDs - their name highlights the special considerations we give to material systems in applied electromagnetics towards lowering loss and reducing system complexity. Utilizing EFCDs at the RF device-level produced the first circuit-type integration of this reconfiguration system; this is identified as the coaxial stub microfluidic impedance transformer (COSMIX). The COSMIX is a small hollowed segment of transmission line with results showing a full reactive loop (capacitive to inductive tuning) around the Smith chart over a 1.2 GHz bandwidth. A second microfluidic application demonstrates a novel antenna reconfiguration mechanism for a 3 GHz microstrip patch antenna. Results showed a 300 MHz downward frequency shift by dielectric colloidal dispersions. Magnetic material produced a 40 MHz frequency shift. The final application demonstrates the dynamically altering microfluidic system for a 3 GHz 1x2 array of linearly polarized microstrip patch antennas. The parallel microfluidic capillaries were imbedded in polydimethylsiloxane (PDMS). Both E- and H-plane designs showed a 250 MHz frequency shift by dielectric colloidal dispersions. Results showed a strong correlation between decreasing electrical length of the elements and an increase of the volume fraction, causing frequency to decrease and mutual coupling to increase. Measured, modeled, and analytical results for impedance, voltage standing wave ratio (VSWR), and radiation behavior (where applicable) are provided.

Communication and reconnaissance systems are requiring increasing flexibility concerning functionality and efficiency for multiband and broadband frequency applications. Circuit-based reconfiguration mechanisms continue to promote radio frequency (RF) application flexibility; however, increasing limitations have resulted in hindering performance. Therefore, the implementation of a "wireless" reconfiguration mechanism provides the required agility and amicability for microwave circuits and antennas without local overhead. The wireless reconfiguration mechanism in this thesis integrates dynamic, fluidic-based material systems to achieve electromagnetic agility and reduce the need for "wired" reconfiguration technologies. The dynamic material system component has become known as electromagnetically functionalized colloidal dispersions (EFCDs). In a microfluidic reconfiguration system, they provide electromagnetic agility by altering the colloidal volume fraction of EFCDs - their name highlights the special considerations we give to material systems in applied electromagnetics towards lowering loss and reducing system complexity. Utilizing EFCDs at the RF device-level produced the first circuit-type integration of this reconfiguration system; this is identified as the coaxial stub microfluidic impedance transformer (COSMIX). The COSMIX is a small hollowed segment of transmission line with results showing a full reactive loop (capacitive to inductive tuning) around the Smith chart over a 1.2 GHz bandwidth. A second microfluidic application demonstrates a novel antenna reconfiguration mechanism for a 3 GHz microstrip patch antenna. Results showed a 300 MHz downward frequency shift by dielectric colloidal dispersions. Magnetic material produced a 40 MHz frequency shift. The final application demonstrates the dynamically altering microfluidic system for a 3 GHz 1x2 array of linearly polarized microstrip patch antennas. The parallel microfluidic capillaries were imbedded in polydimethylsiloxane (PDMS). Both E- and H-plane designs showed a 250 MHz frequency shift by dielectric colloidal dispersions. Results showed a strong correlation between decreasing electrical length of the elements and an increase of the volume fraction, causing frequency to decrease and mutual coupling to increase. Measured, modeled, and analytical results for impedance, voltage standing wave ratio (VSWR), and radiation behavior (where applicable) are provided.

The Archimedean spiral antenna is well-known for its broadband characteristics with circular polarization and has been investigated for several decades. Since their development in the late 1950's, establishing an analytical expression for the characteristics of spiral antenna has remained somewhat elusive. This has been studied qualitatively and evaluated using numerical and experimental techniques with some success, but many of these methods are not convenient in the design process since they do not impart any physical insight into the effect each design parameter has on the overall operation of the spiral antenna. This work examines the operation of spiral antennas and obtains a closed-form analytical solution by conformal mapping and transmission line model with high precision in a wide frequency band. Based on the analysis of spiral antenna, we propose two novel design processes for the stripline-fed Archimedean spiral antenna. This includes a stripline feed network integrated into one of the spiral arms and a broadband tapered impedance transformer that is conformal to the spiral topology for impedance matching the nominally-high input impedance of the spiral. A Dyson-style balun located at the center facilitates the transition between guided stripline and radiating spiral modes. Measured and simulated results for a probe-fed design operating from 2 GHz to over 20 GHz are in excellent agreements to illustrate the synthesis and performance of a demonstration antenna. The research in this work also provides the possibility to achieve conformal integration and planar structural multi-functionality for an Unmanned Air Vehicle (UAV) with band coverage across HF, UHF, and VHF. The proposed conformal mapping analysis can also be applied on periodic coplanar waveguides for integrated circuit applications.

Decades of efforts have pushed the replacement of electrical interconnects by optical links to the interconnects between computers, racks and circuit boards. It may be expected that optical solutions will further be used for inter-chip and intra-chip interconnects with potential benefits in bandwidth, capacity, delay, power consumption and crosstalk. Silicon integration is emerging to be the best candidate nowadays due to not only the dominant status of silicon in microelectronics but also the great advantages brought to the photonic integrated circuits (PICs). Regarding the recent breakthroughs concerning active devices on silicon substrate, the question left is no longer the feasibility of the optical interconnects based on silicon but the competitiveness of the silicon device compared with other alternatives. This thesis focuses on the study of two key components for the optical interconnects, both especially designed and fabricated for silicon platform. One is a high speed electroabsorption modulator (EAM), realized by transferring an InP-based segmented design to the hybrid silicon evanescent platform. The purpose here is to increase the speed of the silicon PICs to over 50  Gb/s or more. The other one is a high performance grating coupler, with the purpose to improve the optical interface between the silicon PICs and the outside fiber-based communication system. An general approach based on the transmission line analysis has been developed to evaluate the modulation response of an EAM with a lumped, traveling-wave, segmented or capacitively-loaded configuration. A genetic algorithm is used to optimize its configuration. This method has been applied to the design of the EAMs on hybrid silicon evanescent platform. Based on the comparison of various electrode design, segmented configuration is adopted for the target of a bandwidth over 40 GHz with as low as possible voltage and high extinction ratio. In addition to the common periodic analysis, the grating coupler is analyzed by the antenna theory assisted with an improved volume-current method, where the directionality of a grating coupler can be obtained analytically. In order to improve the performance of the grating coupler, a direct way is to address its shortcoming by e.g. increasing the coupling efficiency. For this reason, a nonuniform grating coupler with apodized grooves has been developed with a coupling efficiency of 64%, nearly a double of a standard one. Another way is to add more functionalities to the grating coupler. To do this, a polarization beam splitter (PBS) based on a bidirectional grating coupler has been proposed and experimentally demonstrated. An extinction ratio of around -20 dB, as well as a maximum coupling efficiency of over 50% for both polarizations, is achieved by such a PBS with a Bragg reflector underneath. / QC 20100906

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Na prática corrente do projeto de torres de aço treliçadas utilizadas para suportar linhas de transmissão de energia elétrica, a avaliação do comportamento dinâmico das estruturas, de maneira geral, não é considerada. Contudo, o principal carregamento a ser considerado na análise estrutural das torres de transmissão de energia elétrica é produzido pelo vento, que atua de forma dinâmica sobre o sistema estrutural formado pelos cabos e pelas torres. Além disso, muitas estruturas desse tipo apresentam características dinâmicas desfavoráveis, o que as tornam mais vulneráveis a ação do vento.Tendo em vista que muitos acidentes envolvendo torres desse tipo ocorrem ainda que a velocidade de vento utilizada no projeto não tenha sido atingida, é possível que em muitos casos o colapso tenha sido governado pelas ações dinâmicas. Assim sendo, esse estudo tem por objetivo a utilização de uma metodologia de analise estrutural que possibilite uma avaliação mais completa acerca do comportamento das torres de transmissão de energia, considerando, obviamente, as características dinâmicas do sistema. Este trabalho de pesquisa utiliza um modelo de elementos finitos capaz de reproduzir com fidelidade o comportamento acoplado entre os cabos da linha de transmissão e estrutura quando submetidos aos carregamentos dinâmicos e não determinísticos produzidos pelo vento. O modelo tridimensional estudado é constituído por elementos finitos de pórtico e treliça espacial e considera o efeito de não-linearidade geométrica decorrente, principalmente, dos grandes deslocamentos sofridos pelos cabos e isoladores. O carregamento do vento é modelado como um processo aleatório a partir das suas propriedades estatísticas. Os resultados obtidos ao longo desse estudo mostram que a parcela dinâmica da resposta das estruturas pode ser determinante no seu comportamento. Nesse caso, a utilização de uma análise estrutural estática pode resultar no mau dimensionamento das torres e, conseqüentemente, em possíveis acidentes. / In the current design process of steel latticed towers used to support electrical transmission lines, the structures dynamic behaviour is not considered. However, the main loading to be taken into account in the electrical transmission line towers structural analysis is produced by the wind, which acts dynamically over the structural system composed by towers and cables. In addition, its not uncommon for slender towers to present disadvantageous dynamic properties, making them vulnerable to the wind action. Considering that many accidents associated to this kind of structure occur even for wind velocities below that specified in the design process, its possible that most of these accidents have been produced by dynamic actions. Therefore, this study aims to develop a structural analysis method that allows for a more accurate evaluation of the electrical transmission towers behaviour, considering its intrinsic dynamic effects. This study proposed the use of a finite element model that can accurately simulate the coupled behaviour between the transmission line cables and the suspension structures when subjected to the dynamic random loading produced by the wind. The proposed three-dimensional model is composed by beam and truss spatial finite elements and accounts for the geometric non-linearity effects produced by the large displacements associated to the line cables and insulators. The wind loading is modelled as a random process based on its statistical properties. Results obtained throughout this study have indicated that the dynamic response can be very meaningful for the system structural behaviour. In this case, the use of a static structural analysis can lead to a non-trustable design of the towers and, consequently, to potential structural failures.

L’objectif de ce travail est de concevoir et de caractériser un front-end millimétriqueutilisant des lignes de propagation à ondes lentes S-CPW optimisées en technologies CMOS avancées.Ces lignes présentant des facteurs de qualité 2 à 3 fois supérieurs à ceux des lignes classiques de typemicroruban ou CPW.Dans le premier chapitre, l’impact de l’évolution des noeuds technologiques CMOS sur lesperformances des transistors MOS aux fréquences millimétriques et sur les lignes de propagation ainsiqu’un état de l’art concernant les performances des front-end sont présentés. Le deuxième chapitreconcerne la réalisation des lignes S-CPW dans différentes technologies CMOS et la validation d’unmodèle phénoménologique électrique équivalent. Le troisième chapitre est dédié à la conceptiond’amplificateurs de puissance à 60 GHz utilisant ces lignes S-CPW en technologies CMOS 45 et65 nm. Cette étude a permis de mettre en évidence l’apport des lignes à ondes lentes aux performancesdes amplificateurs de puissance fonctionnant dans la gamme des fréquences millimétriques. Uneméthode de conception basée sur les règles d’électro-migration et permettant une optimisation desperformances a été développée. Finalement, un amplificateur faible bruit et un commutateur d’antennetravaillant à 60 GHz et à base de lignes S-CPW ont été conçus en technologie CMOS 65 nm afin degénéraliser l’impact de ce type de lignes sur les performances des front-end millimétriques. / The objective of this work is to design and characterize a millimeter-wave front-end usingthe optimized slow-wave transmission lines S-CPW in advanced CMOS technologies. The qualityfactor of these transmission lines is twice to three times higher than that of the conventionaltransmission lines such as microstrip lines and coplanar waveguides.In the first chapter, the influence of CMOS scaling-down on the performance of transistors atmillimeter-wave frequencies and on the transmission lines was studied. In addition, a state of the artwith regard to the performance of the front-end was presented. The second chapter concerns about therealization of the S-CPW lines in different CMOS technologies and the validation of an electricalequivalent model. The third chapter is dedicated to the design of 60-GHz power amplifiers using theseS-CPW lines in CMOS 45 and 65 nm technologies. This study highlighted the performanceenhancement of power amplifiers operating at millimeter-wave frequencies by using the slow-wavetransmission lines. A design method based on the electro-migration rules was also developed. Finally,a low noise amplifier and an antenna switch operating at 60 GHz were designed in CMOS 65 nm inorder to generalize the impact of such transmission lines on the performance of the millimeter-wavefront-end.

Orientador: Luiz Fernando Bovolato / Banca: Sérgio Kurokawa / Banca: Afonso José do Prado / Resumo: Este trabalho analisa a aplicação em linhas de transmissão do método dos elementos de contorno para cálculo de potencial e campo elétrico, com um enfoque em eletrostática. O método dos elementos de contorno baseia-se numa formulação integral que elimina a discretização do domínio, restando apenas o contorno, permitindo o cálculo do potencial e do campo elétrico no contorno e na região estudada. O trabalho configura-se como uma revisão sobre eletrostática, ressaltando as equações de Laplace e Poisson, que serão utilizadas para encontrar as equações integrais do contorno. Há também vários tópicos relacionados ao campo elétrico de linhas de transmissão, bem como, ás normas brasileiras e recomendações internacionais que devem ser utilizadas no projeto de linhas de transmissão. O método dos elementos de contorno utiliza tais equações integrais para encontrar o potencial e o campo no contorno, e após conhecidos o potencial e o campo no contorno, pode-se aplicar o método em todo o domínio, obtendo o potencial e o campo. Para isso, apenas o contorno do domínio de interesse deve ser discretizado, o que trás uma enorme vantagem sobre os métodos que utilizam formulação diferencial. Neste trabalho, serão descritas as principais características do código computacional desenvolvido e suas sub-rotinas mais importantes. Para validar o programa, os resultados serão comparados com aqueles calculados por um procedimento analítico, sendo mostrada a eficiência da discretização do solo. São apresentados os resultados obtidos da análise do campo elétrico gerado por algumas silhuetas de linhas de transmissão. Os valores do campo elétrico gerado por estruturas compactas são comparados com estruturas convencionais e estruturas reduzidas (semi-compactas), também serão comparados os valores do gradiente de potencial na superfície dos condutores e suas capacitâncias equivalentes. / Abstract: This paper analyses the application in transmission lines of the Boundary Element Method (BEM) of the calculation of potential and electric field, with a focus on electrostatic. The Boundary Element Method is based on an integral formulation that eliminates the discretisation of the domain, remaining only the contour, allowing the calculation of the potential and the electric field in the contour and in the region studied. The work is configured as revision on electrostatic, underscoring the equations of Laplace and Poisson, which will be used to find the integral equations of the contour. There are also several topics related to the electric field of transmission lines, as well as to the standards Brazilian and international recommendations to be used in the design of transmission lines. The Boundary Element Method uses such integral equations for finding the potential and electric field in the contour, and after having known the potential and electric field in the contour, the BEM can be applied in the whole domain, and getting the potential and electric field. Therefore, only the contours of the domain of interest should just be discretized, which backward an enormous advantage on the methods that use formulation differential. This paper will describe the main characteristics of computer code developed and their sub-routines more important. To validate the program, the results will be compared with those calculated by an analytic procedure, being shown the efficiency of discretisation of the soil. The results obtained from analysis of the electric field generated by some silhouettes of transmission lines are presented. The values of the electric field generated by compact structures are compared with conventional structures and reduced structures, also will be compared the values of the gradient of potential on the surface of the conductors and their equivalents capacitances. / Mestre