In order to inject solar power to the utility grid, among various types of inverters, Grid Connected Transformerless Solar Inverters (GCTSI) are mostly preferred for residential or commercial applications. This preference is because of the high energy efficiency and low cost due to the absence of a line frequency or a high frequency transformer. Peak value of the efficiency characteristics of GCTSIs can reach 98%, which are selected topology, component optimization, switching strategy and operating condition dependent. In spite of the attractive energy efficiency characteristics of GCTSIs, due to the lack of galvanic isolation, these inverters are vulnerable to leakage currents, which are prohibitive for the safety and the maintenance reasons. The purpose of this research is to analyze GCTSIs in terms of their leakage current and energy efficiency characteristics. In the research, the leakage current mechanisms of GCTSIs are identified and grid connected solar inverters are classified in terms of their leakage current characteristics including the GCTSIs. In addition to the existing ones, several novel topologies are proposed enriching the family of GCTSIs. The leakage current and the inductor current ripple performances of GCTSI topologies are analyzed and evaluated by detailed simulations for 3 kVA and 10 kVA single-phase systems. In addition, the energy efficiency characteristics of GCTSIs are investigated in these power levels by making use of Calculated Average Power Per Switching Cycle (CAPPSC) method. The efficiency studies with CAPPSC method provide design guidelines and comparison of the GCTSI topologies in terms of their energy efficiency characteristics.

<div>The high-frequency switching of transistors in power electronic (PE) converters is known to cause unintended common-mode (CM) current that flows through parasitically-coupled ground paths. One way to model these currents is to utilize time-domain simulations that capture switching dynamics and the corresponding parasitic response. Although potentially useful, the small time steps required can create a computational burden and limit the usefulness of the approach. In addition, access to internal hardware needed to characterize parasitic parameters is often limited.</div><div><br></div><div>In this thesis, frequency-domain Thevenin equivalent circuits (TECs) are derived to model the CM behavior of PE converters. To do so, periodic linear time-varying (PLTV) analysis is used to develop Thevenin-like models that account for switching behavior of PE circuits. Subsequently, it is shown that in many applications these PLTV TECs can be reduced to traditional linear time-invariant (LTI) forms. Methods to experimentally characterize LTI TEC parameters and couple multiple TECs together for system-level analysis are then established. Finally, the TEC approach is extended to model converters in which common- and differential-mode (CM/DM) behavior are strongly coupled. Simulation and experimental results are used to validate the proposed TEC techniques.</div>

With worldwide growing demand for electric energy, there has been a great interest in exploring photovoltaic (PV) sources. For the PV generation system, the power converter is the most essential part for the efficiency and function performance. In recent years, there have been quite a few new transformerless PV inverters topologies, which eliminate the traditional line frequency transformers to achieve lower cost and higher efficiency, and maintain lower leakage current as well. With an overview of the state-of-the-art transformerless PV inverters, a new inverter technology is summarized in the Chapter 2, which is named V-NPC inverter technology. Based this V-NPC technology, a family of high efficiency transformerless inverters are proposed and detailly analyzed. The experimental results demonstrate the validity of V-NPC technology and high performance of the transformerless inverters. For the lower power level transformerless inverters, most of the innovative topologies try to use super junction metal oxide semiconductor field effect transistor(MOSFET) to boost efficiency, but these MOSFET based inverter topologies suffer from one or more of these drawbacks: MOSFET failure risk from body diode reverse recovery, increased conduction losses due to more devices, or low magnetics utilization. By splitting the conventional MOSFET based phase leg with an optimized inductor, Chapter 3 proposes a novel MOSFET based phase leg configuration to minimize these drawbacks. Based on the proposed phase leg configuration, a high efficiency single-phase MOSFET transformerless inverter is presented for the PV micro-inverter applications. The PWM modulation and circuit operation principle are then described. The common mode and differential mode voltage model is then presented and analyzed for circuit design. Experimental results of a 250 W hardware prototype are shown to demonstrate the merits of the proposed MOSFET based phase-le and the proposed transformerless inverter. New codes require PV inverters to provide system regulation and service to improve the distribution system stabilization. One obvious impact on PV inverters is that they now need to have reactive power generation capability. The Chapter 4 improves the MOFET based transformerless inverter in the Chapter 3 and proposed a novel pulse width modulation (PWM) method for reactive power generation. The ground loop voltage of this inverter under the proposed PWM method is also derived with common mode and differential mode circuit analyses, which indicate that high-frequency voltage component can be minimized with symmetrical design of inductors. A 250-W inverter hardware prototype has been designed and fabricated. Steady state and transient operating conditions are tested to demonstrate the validity of improved inverter and proposed PWM method for reactive power generation, high efficiency of the inverter circuit, and the high-frequency-free ground loop voltage. Besides the high efficiency inverter circuit, the grid connection function is also the essential part of the PV system. The Chapter 5 present the overall function blocks for a grid-connected PV inverter system. The current control and voltage control loop is then analyzed, modeled, and designed. The dynamic reactive power generation is also realized in the control system. The new PLL method for the grid frequency/voltage disturbance is also realized and demonstrate the validity of the detection and protection capability for the voltage/frequency disturbance. At last, a brief conclusion is given in the Chapter 6 about each work. After that, future works on device packaging, system integration, innovation on inverter circuit, and standard compliance are discussed. / Ph. D.

Electromagnetic Interference (EMI), primarily common mode (CM), is problematic in a wide range of electronic circuits due to its propensity to radiate, particularly in high power applications. It is routine for much effort and resources to be dedicated to its characterization and reduction as EMI compliance is a requirement for most electronic systems and devices, including power electronics. Many well-known factors contribute to a system’s EMI performance including intentional coupling from system components as well as unintentional coupling from parasitics. Sources of intentional coupling may include Y-capacitors intended to mitigate EMI as part of a filter. Unintentional coupling is more elusive and can exist throughout the system in PCB layout, cabling, load construction, and internal to components such as inverter bridges. Lesser-known contributions to EMI performance irregularities can be EMI filter asymmetries, switching asymmetries, line impedance variances, and galvanic coupling from the metrology intended to measure EMI. It is critical to understand these contributors to facilitate designs with optimal EMI performance. EMI filters are often added to designs with no consideration to asymmetries in construction and component tolerances. This proposal evaluates the impact to CM currents in cases of coupling or leakage inductance imbalances of a CM choke. Similarly, CM currents are also evaluated for cases when EMI filter Y-capacitor imbalances span the components tolerance band. Also analyzed are switching asymmetries in a typical converter topology to understand EMI impact and evaluate potential benefits if intentional asymmetric switching is applied. A practical method is introduced to measure line impedance upstream of devices under test as line impedance variation can impact the performance of EMI filter design. However, few documented practices exist to measure line impedance without specialized instrumentation. Finally, this work proposes a streamlined method for conducted emissions evaluation employing an oscilloscope, differential voltage probes, and post-processing software implemented in MATLAB. This method eliminates unintended metrology ground coupling that can significantly impact EMI measurements and minimizes risk of instrumentation damage particularly in high power systems.

Due to modularity and easy scalability, modular multilevel converters (MMCs) are deemed the most suitable for high-voltage and medium-voltage power conversion applications. However, large module capacitors are usually required in MMCs to store large circulating power of line-frequency and its harmonics that flow through the capacitors. Even though several methods for minimizing the circulating power have been proposed in the literature, there is still the need for a systematic and simplified approach of addressing these control strategies and evaluating their efficacy. Moreover, the generally accepted feedback control architecture for the MMC is complicated, derived through a rigorous mathematical analysis, and therefore, not easy to intuitively comprehend. Recently, a method of modeling of the MMC based on state-plane analysis and coordinate transformation, is proposed in the literature. Based on the state-plane analysis, two kinds of circulating power in the MMC are identified that are orthogonal to each other. This means these two circulating power can be controlled individually without affecting each other. To control these circulating power, in the literature, a decoupled equivalent circuit model is developed through the coordinate transformation which clearly suggests a means for minimizing these circulating power. Further extending this work, in this thesis, the existing control concepts for reducing the circulating power are unveiled in a systematic and simplified manner utilizing the decoupled equivalent circuit model. A graphical visualization of circulating power using the state-planes is provided for each control strategy to readily compare its efficacy. Moreover, the generally accepted control architecture of the MMC is presented in an intuitive and simplified way using the decoupled circuit model. The important physics related to control implementation, originally hidden behind the complicated mathematics, is explained in detail. / Master of Science / A power converter is an electrical device that converts electrical energy from one form to another in order to be compatible with the load demand. A typical power converter consists of semiconductor switches, inductor, capacitor etc. These power converters are required in a wide range of applications: automotive and traction, motor drives, renewable energy conversion, energy storage, aircraft, power generation, transmission, and distribution, to name a few. Many of these applications are continuously increasing their power capacity to handle the escalating demands of energy that exist due to rising population numbers, industrialization, urbanization etc. Consequently, it has been a responsibility of power electronics engineers and researchers to develop power converters that can handle high voltages and high currents. Multilevel power converters have been the key-enabling developments that can withstand high-voltages while using traditional low-voltage semiconductor switches. Several multilevel converters such as the neutral point clamped converter, flying capacitor converter, cascaded H-bridge converter, modular multilevel converter (MMC) etc. have been developed and commercialized in the last two decades. Among them, the MMC is a widely accepted topology for medium- and high-voltage power conversion applications. In an MMC, several modules are stacked together in series, and each module consists of semiconductor switches and a capacitor. The series connection of the modules enables the MMC to handle high-voltage power conversion using low-voltage traditional semiconductor switches. The voltage rating of an MMC can be easily scaled-up by simply increasing the number of modules in each arm. Moreover, since several identical modules are connected in each arm, the structure of the MMC is highly modular which helps greatly in manufacturing and design. Nonetheless, in MMCs, generally large circulating power flow to the capacitor in each module, which leads to significant voltage ripples. To suppress these voltage ripples, a large capacitor is required in each module, leading to large size and weight of the converter. In the literature, several control strategies have been proposed to minimize the circulating power. However, there is still the need for a systematic and simplified approach of addressing these control strategies and evaluating their efficacy. Moreover, the generally accepted feedback control architecture for the MMC is complicated, derived through a rigorous mathematical analysis, and therefore, not easy to intuitively comprehend. Recently, a decoupled equivalent circuit model has been developed in the literature. This model clearly explains the process of power flow in the MMC between input and output and the nature of the circulating power. The equivalent circuit model provides the circulating power, that are orthogonal to each other, meaning they can be controlled individually without affecting each other. Moreover, the equivalent circuit model clearly suggests a means for minimize the circulating power by providing two "ideal" control laws. Further extending this work, in this thesis, the existing control concepts for reducing the circulating power are unveiled in a systematic and simplified manner utilizing the decoupled equivalent circuit model. Moreover, the generally accepted control architecture of the MMC is presented in an intuitive and simplified way via the decoupled circuit model. The important physics related to control implementation, originally hidden behind the complicated mathematics, is explained in detail.

Thesis (DPhil (Electrical and Electronic Engineering))--University of Stellenbosch, 2011. / ENGLISH ABSTRACT: The Karoo Array Telescope (KAT) will be used by South Africa in its bid to host the international Square Kilometre Array (SKA). As the SKA will have orders of magnitude greater sensitivity than existing radio telescopes, it will also be concomitantly more sensitive to radio frequency interference (RFI). The influence of RFI on a differential mode (DM) KAT conductive system is an unavoidable phenomenon. In this context, the conductive or galvanic system can, in its most basic form, consist of a source, a load, and connecting conductors. It can also, in the case of the KAT-7 seven-dish interferometer, consist of each telescope, its functional cabling, and the main correlator connecting the telescopes together. However, additional connections between the system and the environment exist. These might be an intentional connection made to the earthing layout, or unintentional connections due to parasitic capacitances and inductive connections. As a result of this, additional conductive systems are created which carry common mode (CM) currents. Interference present in such CM paths enters the DM system through the transfer impedance (Zt) between them. To reduce or eliminate this interference in the DM system, Zt has to be minimised. The use of an earthed parallel conductor (EPC as commonly referred to) in the form of a cable tray is considered to be one of the principal methods to reduce Zt. The properties of cable trays as EPCs at wavelengths which are greater than the tray length are well documented. One main focus in this dissertation is on cable tray mid-span and end connections. They are not well described in the literature over the wide range of frequencies that is expected for KAT developments. The influence of the most common connections on the measured and computed Zt of the cable tray is determined. Computer Simulation Technology’s Microwave Studio (CST MWS) is employed to validate the measured results and also to enable visualisation of the fields and currents. Recommendations for the best connection to use for any cable tray installation is given. The overall shielding ability of optimally-connected cable trays has been evaluated using a physical and computational model. In both cases the induced voltage on a victim conductor, and far-field gain functions for varying angles of incidence onto the model, are determined. The results also show that for certain scenarios, most of the coupling to the victim conductor, takes place inside the end enclosure and not the cable tray. In general, properly-connected cable trays do provide protection to their enclosed conductors, even at frequencies were the wavelength is much shorter than the width of the tray. The second main focus arises from an on-site radio frequency (RF) current audit undertaken on two of the seven KAT-7 telescopes. Shielding measures, such as interface barriers at the floor and roof of the lower telescope pedestal, are studied. The investigation is facilitated by the development of an accurate physical and computational scale model of the dish. Direct current injection and plane wave illumination methods are used to excite the system. The measured CM current distributions are compared and comments made regarding the validity of the measurement procedure. The CM currents, measured around the outside of the lower pedestal show higher levels when a direct current path to ground is established, as opposed to when no clear path exists. This finding suggests at least two methods of preventing CM interference entering or leaving the pedestal: harden the floor and roof barriers, or manage current paths outside the telescope. Related to this, CM currents measured either side of the telescope interface barriers, are used to determine the level of shielding the interface provides. When compared to the common definition of shielding effectiveness, the current measurement provides more conservative shielding estimates. The research in this dissertation has influenced, and will continue to influence, the layout of galvanic systems for the present KAT-7 structures and the anticipated developments to MeerKAT. / AFRIKAANSE OPSOMMING: Die Karoo Array Telescop (KAT) sal gebruik word deur Suid-Afrika in sy poging om die bod te kry om die internasionale Square Kilometre Array (SKA) te huisves. Aangesien die SKA ordes meer sensitief sal wees as bestaande radioteleskope, sal dit terselfdertyd ook meer sensitief wees vir radio frekwensie steurnisse. Die invloed van radio frekwensie steurnisse op die differentiëlemodus KAT geleidende netwerk is iets onvermydeliks. Binne hierdie konteks kan die geleidende netwerk, in sy mees basiese vorm, bestaan uit ʼn bron, ʼn las, en verbindingsgeleiers. Dit kan ook, in die geval van die KAT-7 sewe-teleskoop interferometer, bestaan uit elke teleskoop, sy funksionele bekabeling, en die korrelator wat die verbinding is tussen al die teleskope. Daar is egter, addisionele verbindings tussen die netwerk en die onmiddellike omgewing rondom dit. Hierdie verbindings kan opsetlik gemaak word deur byvoorbeeld, verbinding van die netwerk se aardkabel aan die res van die beaarding, of onopsetlik deur parasitiese kapasitansies en induktiewe verbindings. As gevolg hiervan word addisionele geledende netwerke geskep waarin gemenemodus strome kan vloei. Indien daar wel strome in die gemenemodus netwerk vloei, word dit oorgedra aan die differentiëlemodus netwerk deur ʼn oordragimpedansie (Zt) wat teenwoordig is tussen die twee. Om dus die steurnisse in die differentiëlemodus netwerk te verminder of te elimineer, moet Zt tot ʼn minimum beperk word. Die gebruik van ʼn geaarde parallelle geleier in die vorm van ʼn kabelkanaal, word beskou as een van die mees doeltreffendste metodes om Zt te verminder. Die eienskappe van kabelkanale as geaarde parallelle geleiers by frekwensies waar die golflengtes langer is as die van die kabelkanaal, is volledig gedokumenteer. Een van die belangrikste fokuspunte in hierdie verhandeling is rakende die kabelkanale se middel en eindpuntverbindings. Hulle word nie goed beskryf in die beskikbare literatuur nie, en weliswaar vir wyeband doeleindes wat vir KAT ontwikkelinge verwag word. Die invloed van die mees algemeenste kabelkanaal verbindings op gemete en berekende Zt word bepaal. Computer Simulation Technology’s Microwave Studio (CST MWS) word eerstens, gebruik om die akkuraatheid van die gemete resultate te bewys en tweedens, deur visualisering van E-veld en gemenemodus oppervlak strome. Aanbevelings vir die beste verbindings vir enige kabelkanaal opstelling word gegee. Die algemene afskerminsvermoeë van ʼn idiaal-verbinde kabelkanaal word bepaal deur middel van metings en simulasies. In beide gevalle word die geïnduseerde spanning op ʼn slagoffer kabel, en die verveld aanwins funksie bepaal vir verskillende invalshoeke op die model. Die resultate toon verder dat vir spesifieke gevalle wat beskou word, die meeste koppeling binne die kabinet aan die einde van die kabelkanaal plaasvind. Oor die algemeen verskaf goed verbinde kabelkanale wel ʼn sekere vlak van beskerming aan kabels binne die kabelkanaal, selfs by frekwensies waar die golflengte baie korter is as die breedte van die kabelkanaal. Die tweede belangrike fokuspunt spruit voort uit ʼn radio frekwensie stroomoudit, wat twee van die KAT-7 teleskope evalueer het. Afskermingsmatreëls soos die kabelhindernisse op die vloer en dak van die onderste teleskoop voetstuk, word bestudeer. Tesame met die metings op die werklike teleskoop wat geneem is, word ʼn akkurate fisiese en simulasie skaalmodel geskep om die metings beter te analiseer. Direkte stroominspuitings metode en platvlakgolf beligting word gebruik om gemenemodus strome op die teleskoop se struktuur te induseer. Die gemenemodus stroomverspreiding vir beide tegnieke word vergelyk in ʼn poging om kommentaar te lewer rakende die geldigheid van die meettegniek. Die gemenemodus strome wat aan die buitekant van die teleskoop voetstuk gemeet word, is hoër wanneer ʼn direkte stroompad na grond op die voetstuk geskep word in vergelyking met ʼn ongedefinieerde pad. Hierdie verskynsel dui daarop dat ten minste twee metodes bestaan om die ongevraagde gemenemodus strome te verhoed om aan die binnekant van die teleskoop voetstuk te vloei. Die een is die verbetering van die vloer en dak kabelhindernisse, en die ander is verbetering van die stroompad (stroompaaie) aan die buitekant van die voetstuk sodat ʼn meer direkte pad na grond geskep word. Die gemiddelde gemenemodus strome weerskante van die kabelhindernis, kan gebruik word op te bepaal hoeveel afskerming die hindernis bied. Indien die berekende waardes by verskillende frekwensies vergelyk word met die tradisionele filter doeltreffendheid (Zt), word ʼn meer konserwatiewe beraming verkry. Die navorsing in hierdie verhandeling het alreeds, maar sal ook die toekomstige uitleg van galvaniese stelsels vir KAT-7 asook die verwagte MeerKAT beïnvloed.

Ce mémoire présente le travail de thèse réalisé auprès des laboratoires LAPLACE et SATIE ainsi que les services EDYNE3 et EDYYLIC d'AIRBUS OPERATIONS. Le sujet porte sur les problèmes CEM apparaissant dans les convertisseurs de puissance embarqués à bord des futurs avions plus électriques. Le manuscrit est composé de cinq parties. La première partie, d'introduction, traite de la problématique CEM avion, la deuxième de la modélisation des éléments de la chaine de conversion DC/AC étudiée. Le troisième est composé d'une étude comparative par simulation des différentes solutions. La quatrième partie traite de la réalisation de la solution choisie et enfin le cinquième et dernier chapitre de l'étude expérimentale de celle-ci. / The studies conducted during this thesis deals with conducted EMC problems of an inverter associated to its actuator. Accurate high frequency models of every element of the DC/AC converter and actuator have been realised from measures. Then a comparative study of different topologies of converters have been led from simulations in order to determine the best solution minimising EMC current. The selected inverter was realised and the experimental results were compared to simulations validating them. Finally, a comparison of EMC filters architecture led to choose a better solution in order to avoid the increasing of mass.

Nos últimos 20 anos, o inversor de frequência PWM passou a ser largamente utilizado. Dentre as razões principais de sua ampla difusão, cita-se a introdução do IGBT que, devido à sua capacidade de comutação rápida, possibilitou a redução de perdas, tamanho e custo dos conversores, ao mesmo tempo em que permitiu o uso do motor de indução em aplicações que demandam alto desempenho dinâmico. Porém, esta mesma característica que trouxe tais benefícios, contribuiu para acentuar os efeitos relacionados a fenômenos eletromagnéticos, devido à rápida transição que produz sinais em alta frequência, responsáveis por tornar os sistemas de acionamentos potenciais fontes de interferência eletromagnética. Os principais fenômenos desta natureza são a tensão de onda refletida, associada a ruídos de modo diferencial, e a corrente de modo comum. De maneira geral, a corrente de modo comum tem maior potencial de contribuição para a emissão radiada que a de modo diferencial e, em sistemas de acionamento, constitui a principal fonte de interferência eletromagnética. Esta não é resultado apenas da rápida comutação dos IGBTs, mas, também, da configuração do sistema, que propicia o surgimento das tensões de modo comum. Este trabalho visa estudar alternativas que minimizem, naturalmente, estas tensões através de sua configuração, reduzindo-se a necessidade de novos componentes, como filtros passivos ou circuitos ativos. Ressalta-se, assim, que o primeiro passo consistiu-se na compreensão do problema de geração da tensão de modo comum. Visando atingir os propósitos anteriormente mencionados, foi idealizado um inversor, composto de duas pontes chaveando em oposição, chamado de duplo-trifásico. A partir de simulações, avaliou-se seu potencial e definiu-se sua implementação, a partir de dois inversores idênticos. Testou-se o protótipo contra um terceiro módulo de mesmo tipo, em iguais condições. A comparação dos resultados indicou que a configuração proposta pode ser eficiente no controle do ruído de modo comum, obtendo-se valores de redução de 86%. Nota-se que o comportamento observado via o protótipo, permite um melhor entendimento dos fenômenos relativos à circulação da corrente de modo comum. Desta forma, à luz da pesquisa realizada, nota-se que, conceitualmente, a solução proposta se destaca das demais apresentadas na literatura pela sua exequibilidade e simplicidade. / Over the last 20 years, the PWM frequency inverter has become widely used. Among the main reasons for its widespread distribution is the introduction of the IGBT, due to its ability for fast switching allowing the reduction of losses, size and cost of the converters, and the use of the induction motor in applications that demand high dynamic performance. However, this same characteristic that brought such benefits, stressed the effects related to electromagnetic phenomena, due to the fast transition that produces high frequency signals, responsible for turning drive systems, potential sources of electromagnetic interference. The main phenomena of this nature are the reflected wave overvoltage, associated with differential mode noise, and the common mode current. In general, the common mode current has greater potential of contribution to the emission of radiated interference than the differential mode current and is the main source of electromagnetic interference in drive systems. The common mode current is not just the result of the fast transitions of IGBTs, but also of the system configuration, which produces common mode voltages. This work aims to study alternatives that minimize, naturally, those voltages through its configuration, reducing the need for additional components, such as passive filters or active circuits. It is pointed, that the first step consisted in the understanding of the problem of the common mode voltage generation. In order to address this objective, an inverter composed of two bridges, switching in opposition, called double-three-phase was designed. Its potential for noise mitigation, as well, its implementation, with two identical inverters was evaluated by simulations. The prototype was tested against a third module of the same type, under the same conditions. The comparison of the results indicated that the proposed configuration can be efficient in the common mode noise control, with 86% reduction. The behavior observed through the prototype allows a better understanding of the phenomena involving common mode current. Thus, in the light of the research, we note that, conceptually, the proposed solution stands out for its practicality and simplicity, from the others presented in the literature.

Nos últimos 20 anos, o inversor de frequência PWM passou a ser largamente utilizado. Dentre as razões principais de sua ampla difusão, cita-se a introdução do IGBT que, devido à sua capacidade de comutação rápida, possibilitou a redução de perdas, tamanho e custo dos conversores, ao mesmo tempo em que permitiu o uso do motor de indução em aplicações que demandam alto desempenho dinâmico. Porém, esta mesma característica que trouxe tais benefícios, contribuiu para acentuar os efeitos relacionados a fenômenos eletromagnéticos, devido à rápida transição que produz sinais em alta frequência, responsáveis por tornar os sistemas de acionamentos potenciais fontes de interferência eletromagnética. Os principais fenômenos desta natureza são a tensão de onda refletida, associada a ruídos de modo diferencial, e a corrente de modo comum. De maneira geral, a corrente de modo comum tem maior potencial de contribuição para a emissão radiada que a de modo diferencial e, em sistemas de acionamento, constitui a principal fonte de interferência eletromagnética. Esta não é resultado apenas da rápida comutação dos IGBTs, mas, também, da configuração do sistema, que propicia o surgimento das tensões de modo comum. Este trabalho visa estudar alternativas que minimizem, naturalmente, estas tensões através de sua configuração, reduzindo-se a necessidade de novos componentes, como filtros passivos ou circuitos ativos. Ressalta-se, assim, que o primeiro passo consistiu-se na compreensão do problema de geração da tensão de modo comum. Visando atingir os propósitos anteriormente mencionados, foi idealizado um inversor, composto de duas pontes chaveando em oposição, chamado de duplo-trifásico. A partir de simulações, avaliou-se seu potencial e definiu-se sua implementação, a partir de dois inversores idênticos. Testou-se o protótipo contra um terceiro módulo de mesmo tipo, em iguais condições. A comparação dos resultados indicou que a configuração proposta pode ser eficiente no controle do ruído de modo comum, obtendo-se valores de redução de 86%. Nota-se que o comportamento observado via o protótipo, permite um melhor entendimento dos fenômenos relativos à circulação da corrente de modo comum. Desta forma, à luz da pesquisa realizada, nota-se que, conceitualmente, a solução proposta se destaca das demais apresentadas na literatura pela sua exequibilidade e simplicidade. / Over the last 20 years, the PWM frequency inverter has become widely used. Among the main reasons for its widespread distribution is the introduction of the IGBT, due to its ability for fast switching allowing the reduction of losses, size and cost of the converters, and the use of the induction motor in applications that demand high dynamic performance. However, this same characteristic that brought such benefits, stressed the effects related to electromagnetic phenomena, due to the fast transition that produces high frequency signals, responsible for turning drive systems, potential sources of electromagnetic interference. The main phenomena of this nature are the reflected wave overvoltage, associated with differential mode noise, and the common mode current. In general, the common mode current has greater potential of contribution to the emission of radiated interference than the differential mode current and is the main source of electromagnetic interference in drive systems. The common mode current is not just the result of the fast transitions of IGBTs, but also of the system configuration, which produces common mode voltages. This work aims to study alternatives that minimize, naturally, those voltages through its configuration, reducing the need for additional components, such as passive filters or active circuits. It is pointed, that the first step consisted in the understanding of the problem of the common mode voltage generation. In order to address this objective, an inverter composed of two bridges, switching in opposition, called double-three-phase was designed. Its potential for noise mitigation, as well, its implementation, with two identical inverters was evaluated by simulations. The prototype was tested against a third module of the same type, under the same conditions. The comparison of the results indicated that the proposed configuration can be efficient in the common mode noise control, with 86% reduction. The behavior observed through the prototype allows a better understanding of the phenomena involving common mode current. Thus, in the light of the research, we note that, conceptually, the proposed solution stands out for its practicality and simplicity, from the others presented in the literature.

L’intégration monolithique de filtres et de diodes contre les décharges électrostatiques sur silicium est une solution à bas cout et fiable pour protéger les interfaces de transfert de données des appareils nomades. La formation de caissons isolants de silicium poreux sous les filtres permet d’améliorer en partie leur performance. Cette thèse avait pour but de poursuivre l’intégration de démonstrateurs RF sur silicium poreux et de proposer des voies de fonctionnalisation du matériau en vue d’optimiser les caractéristiques des filtres. Tout d’abord, les configurations bénéfiques d’intégration de caissons poreux à un filtre de mode commun (ECMF) ont été étudiées. Ensuite, une optimisation de l’étape d’oxydation post-anodisation a été réalisée afin d’améliorer la qualité et la stabilité de l’isolation électrique. Pour cela, les mécanismes d’oxydation, les propriétés chimiques et les propriétés électriques du silicium mésoporeux oxydé ont été mises en perspective avec la nature du traitement appliqué. Enfin, l’insertion dans les pores de nanoparticules ferromagnétiques de forte perméabilité a été menée dans le but d’augmenter la densité d’inductance par unité de surface. / Monolithic integration of interference filters and protection diodes on silicon is a viable and mature technology used to protect high-speed serial interfaces of nomadic devices. To enhance the filters performance, porous silicon can be formed by anodization specifically underneath the filter area. This thesis aimed to pursue the integration of RF prototypes on porous silicon and also to suggest strategies of material functionalization in order to optimize the filter characteristics. First, various configurations of common-mode filters were integrated on porous silicon and their performances were compared. Then, the post-anodization oxidation step has been optimized in order to provide the most efficient and stable electrical isolation. The oxidation mechanisms were discussed. The surface chemistry of porous silicon and its electrical behavior have been put in perspective with the oxidation treatments. Finally, this work suggested experimental methods to synthesize and characterize a ferromagnetic porous silicon-based nanocomposite for the improvement of the inductance density per unit area.