Avec l'objectif d'améliorer le rendement des lampes à excimères (Excilampe) à décharge à barrière diélectrique (DBD), un convertisseur en mode de courant, qui permet un ajustement précis de la puissance électrique injectée dans ce type des lampes, à été conçu et mis en oeuvre. Ce convertisseur fournit à la lampe un courant de forme d'onde carrée contrôlé au moyen de trois paramètres: l'amplitude, la fréquence et le rapport cyclique, pour obtenir un contrôle total de l'énergie électrique transmise à l'excilampe DBD. La mise en oeuvre intègre un transformateur élévateur comme interface entre la lampe et un commutateur. Les expériences démontrent le principe de fonctionnement de ce convertisseur, y compris les mesures de puissance du rayonnement UV. Les degrés de liberté du convertisseur sont utilisées pour analyser le comportement de la lampe sous différentes combinaisons de ces trois paramètres, et sont utilisés pour déterminer le point de fonctionnement optimal de la lampe. Ensuite, un convertisseur à résonance du type onduleur série, est proposé pour alimenter la lampe avec une grande efficacité électrique. Afin de contrôler effectivement la puissance de la lampe, le mode de fonctionnement de ce convertisseur utilise le mode de conduction discontinue et la commutation douce (ZCS), avec lequel on obtient aussi de faibles émissions électromagnétiques et l'on réduit les pertes de commutation. Les relations mathématiques obtenus à partir de l'analyse du diagramme de phase, ont été validées par des simulations et avec des résultats expérimentaux. Enfin, différentes topologies d'alimentations pour DBD sont comparées analytiquement et expérimentalement pour évaluer objectivement les avantages de chaque approche. Une des perspectives de ce travail est l'application de l'alimentation en créneaux pour l'étude de la performance d'autres types de réacteurs et d'excilampes DBD. / With the aim to provide a scientific tool for the enhancement of the Dielectric Barrier Discharge (DBD) Excimer Lamps (Excilamp) performance, a current-mode converter that allows an accurate adjustment of the electrical power injected into one of those lamps, is designed and implemented. With the proposed converter, the current supplied to the lamp has a square shape, controlled by means of three parameters: amplitude, duty cycle and frequency, which provides full control of the lamp electrical power. Implementation is made considering a step-up transformer interfacing the high-voltage lamp with the converter. Experiments demonstrate the operating principle of this converter, including UV power measurements for a DBD XeCl Excilamp. The capabilities of the converter are used to analyze the lamp behavior under different combinations of these three parameters, illustrating its capabilities for finding the optimal operating point. Then a series-resonant inverter for the supply of DBD) excilamp is proposed. In order to effectively control the lamp power, the operating mode of this converter combines discontinuous current-mode and soft-commutation (ZCS), obtaining as well low electromagnetic emissions, and reduced switching losses. The mathematical relationships obtained from state plane analysis, are validated with simulations and experimental results. Finally, several topologies of DBDs power supplies are compared analytical and experimentally to elucidate the advantages of each approach. After this work, one of the perspectives is the application of the square-shape supply in the performance study of other types of DBD excilamps and DBD reactors.

No description available.

The electrification of transportation is increasingly of interest to governments around the world as a means of contributing to the achievement of climate change goals. Transportation is a significant source of greenhouse gas emissions, but it is also the backbone of the global economy and local mobility. Electrification is widely seen as a promising pathway to reducing greenhouse gas emissions from transportation while continuing to support economic growth. Multiphase machines have distinctive features that draw attention in the transportation electrification domain due to their features. Recently, powertrains based on the current-source inverter (CSI) are getting more attention to be a more reliable structure for Electric Vehicles (EVs) by replacing the dc-link capacitor with a choke inductor. This thesis combines these two technologies to develop a more reliable, compact powertrain for heavy-duty electric vehicles. First, a survey covers the recent advances in several aspects such as topology, control, and performance to evaluate the possibility and the future of exploiting them more in EV applications. The six-phase drives are extensively covered here because of their inherent structure as a dual three-phase system which eases the production process. The survey presents the different topologies used in dual three-phase drives, the modulation techniques used to operate them, the status of using multiphase drives in traction applications industrially, and the upcoming trends toward promoting this technology. New powertrain configurations for heavy-duty electric vehicles (HDEV) are proposed based on current-source inverters (CSI) and asymmetrical six-phase electric machines. Since the six-phase CSI comprises two three-phase CSIs, multiple configurations can arise based on the connection between the two CSIs. In this context, the proposed powertrain configurations are based on parallel, cascaded, and standalone six-phase CSIs. The standalone topology is based on separating the two three-phase converters by supplying each converter with a dedicated dc-dc converter. A new and straightforward method is proposed to extend the six-phase standalone CSI. The proposed technique employs the vector space decomposition (VSD) to mitigate the inverter current harmonics and extend the linear modulation region by about 8%. For motor drive applications, increasing the fundamental output component can reflect higher torque production capability for the same drive size, given that thermal limits are not exceeded. Moreover, to increase the drive's reliability, space vector modulation (SVM) techniques are developed to operate the six-phase CSI while reducing the common-mode voltage (CMV) content associated with the switching of semiconductors. The SVM techniques select the switching states associated with the minimum CMV value offline to eliminate the need for measurements. Experimental validation of the proposed algorithms is presented to operate a scaled-down six-phase PMSM fed by the proposed powertrain configuration. These proposed techniques make the CSI- based powertrain a promising solution for future HDEVs in terms of cost, performance, and reliability. / Thesis / Candidate in Philosophy

This dissertation presents the analysis and development of an innovative semiconductor device and topology for the high power current source converter (CSC). The CSC is very attractive in high power applications due to its lower output dv/dt, easy regeneration capability and implicit short-circuit protection. Traditionally, either a symmetrical gate turn-off (GTO) thyritor or an asymmetrical GTO in series with a diode is used as the power switch in the CSC. Since the GTO has a lower switching speed and requires a complicated gate driver, the symmetrical GTO based CSC usually has low dynamic response speed and low efficiency. To achieve high power rating, fast dynamic response speed and low harmonics, an advanced semiconductor device and topology are needed for the CSC. Based on symmetrical GTO and power MOSFET technologies, a symmetrical emitter turn-off (ETO) thyristor is developed that shows superior switching performance, high power rating and reverse voltage blocking capability. The on-state characteristics, forced turn-on characteristics, forced turn-off characteristics and the load-commutated characteristics are studied. Test results show that although the load-commutation loss is high, the developed symmetrical ETO is suitable for use in high power CSC due to its low conduction loss, fast switching speed and reverse voltage blocking capability. The snubberless turn-on capability is preferred for a semiconductor device in a power conversion system, and can be achieved for devices with forward biased safe operation area (FBSOA). The FBSOA of the ETO is investigated and experimentally demonstrated. The ETO device has excellent FBSOA due to the negative feedback provided by the emitter switch. However, the FBSOA for a large area ETO is poor. A new ETO concept is therefore proposed for future development in order to demonstrate the FBSOA over a large area device. To improve the turn-on performance of the large area ETO, a novel concept, named the transistor-mode turn-on, is proposed and studied. During the transistor-mode turn-on process, the ETO behaves like a transistor instead of a thyristor. Without a snubber, the transistor-mode turn-on for the ETO is hard to achieve. Through the selection of a proper gate drive and di/dt snubber, the transistor-mode turn-on can be implemented, and the turn-on performance for the ETO can be dramatically improved. To increase the power rating of the CSC without degrading the utilization of power semiconductor devices, a novel multilevel CSC, named the parallel-cell multilevel CSC, is proposed. Based on a six-switch CSC cell, the parallel-cell multilevel CSC has the advantages of high power rating, low harmonics, fast dynamic response and modularity. Therefore, it is very suitable for high power applications. The power stage design, modeling, control and switching modulation scheme for a parallel-cell multilevel CSC based static var compensator (STATCOM) are analyzed and verified through simulation. / Ph. D.

Des travaux récents indiquent qu'il existe des différences fondamentales entre les systèmes visuel et auditif: tandis que le premier semble échantillonner le flux d'information en provenance de l'environnement, en passant d'un "instantané" à un autre (créant ainsi des cycles perceptifs), la plupart des expériences destinées à examiner ce phénomène de discrétisation dans le système auditif ont mené à des résultats mitigés. Dans cette thèse, au travers de deux expériences de psychophysique, nous montrons que le sous-échantillonnage de l'information à l'entrée des systèmes perceptifs est en effet plus destructif pour l'audition que pour la vision. Cependant, nous révélons que des cycles perceptifs dans le système auditif pourraient exister à un niveau élevé du traitement de l'information. En outre, nos résultats suggèrent que du fait des fluctuations rapides du flot des sons en provenance de l'environnement, le système auditif tend à avoir son activité alignée sur la structure rythmique de ce flux. En synchronisant la phase des oscillations neuronales, elles-mêmes correspondant à différents états d'excitabilité, le système auditif pourrait optimiser activement le moment d'arrivée de ses "instantanés" et ainsi favoriser le traitement des informations pertinentes par rapport aux événements de moindre importance. Non seulement nos résultats montrent que cet entrainement de la phase des oscillations neuronales a des conséquences importantes sur la façon dont sont perçus deux flux auditifs présentés simultanément ; mais de plus, ils démontrent que l'entraînement de phase par un flux langagier inclut des mécanismes de haut niveau. Dans ce but, nous avons créé des stimuli parole/bruit dans lesquels les fluctuations de l'amplitude et du contenu spectral de la parole ont été enlevés, tout en conservant l'information phonétique et l'intelligibilité. Leur utilisation nous a permis de démontrer, au travers de plusieurs expériences, que le système auditif se synchronise à ces stimuli. Plus précisément, la perception, estimée par la détection d'un clic intégré dans les stimuli parole/bruit, et les oscillations neuronales, mesurées par Electroencéphalographie chez l'humain et à l'aide d'enregistrements intracrâniens dans le cortex auditif chez le singe, suivent la rythmique "de haut niveau" liée à la parole. En résumé, les résultats présentés ici suggèrent que les oscillations neuronales sont un mécanisme important pour la discrétisation des informations en provenance de l'environnement en vue de leur traitement par le cerveau, non seulement dans la vision, mais aussi dans l'audition. Pourtant, il semble exister des différences fondamentales entre les deux systèmes: contrairement au système visuel, il est essentiel pour le système auditif de se synchroniser (par entraînement de phase) à son environnement, avec un échantillonnage du flux des informations vraisemblablement réalisé à un niveau hiérarchique élevé. / Recent research indicates fundamental differences between the auditory and visual systems: Whereas the visual system seems to sample its environment, cycling between "snapshots" at discrete moments in time (creating perceptual cycles), most attempts at discovering discrete perception in the auditory system failed. Here, we show in two psychophysical experiments that subsampling the very input to the visual and auditory systems is indeed more disruptive for audition; however, the existence of perceptual cycles in the auditory system is possible if they operate on a relatively high level of auditory processing. Moreover, we suggest that the auditory system, due to the rapidly fluctuating nature of its input, might rely to a particularly strong degree on phase entrainment, the alignment between neural activity and the rhythmic structure of its input: By using the low and high excitability phases of neural oscillations, the auditory system might actively control the timing of its "snapshots" and thereby amplify relevant information whereas irrelevant events are suppressed. Not only do our results suggest that the oscillatory phase has important consequences on how simultaneous auditory inputs are perceived; additionally, we can show that phase entrainment to speech sound does entail an active high-level mechanism. We do so by using specifically constructed speech/noise sounds in which fluctuations in low-level features (amplitude and spectral content) of speech have been removed, but intelligibility and high-level features (including, but not restricted to phonetic information) have been conserved. We demonstrate, in several experiments, that the auditory system can entrain to these stimuli, as both perception (the detection of a click embedded in the speech/noise stimuli) and neural oscillations (measured with electroencephalography, EEG, and in intracranial recordings in primary auditory cortex of the monkey) follow the conserved "high-level" rhythm of speech. Taken together, the results presented here suggest that, not only in vision, but also in audition, neural oscillations are an important tool for the discretization and processing of the brain's input. However, there seem to be fundamental differences between the two systems: In contrast to the visual system, it is critical for the auditory system to adapt (via phase entrainment) to its environment, and input subsampling is done most likely on a hierarchically high level of stimulus processing.

This thesis describes the generic modeling of a Doubly-Fed Induction Generator (DFIG) based wind turbine. The model can also represent a wind plant with a group of similar wind turbines lumped together. The model is represented as a controlled current source which injects the current needed by the grid to supply the demanded real and reactive power. The DFIG theory is explained in detail as is the rationale for representing it by a regulated current source. The complete model is then developed in the time-domain and phasor domain by the interconnection of various sub-systems, the functions of which have been described in detail. The performance of the model is then tested for steady-state and dynamic operation. The model developed can be used for bulk power system studies and transient stability analysis of the transmission system. This thesis uses as its basis a report written for NREL [1]. / text

In conventional three-phase PWM (Pulse Width Modulation) current source converter based STATCOM (Static Synchronous Compensator) applications, DC link current is kept constant at a predefined value and the reactive power of STATCOM is controlled by varying modulation index. This control strategy causes unnecessary power losses especially when the reactive power of STATCOM is low. For this purpose, in order to reduce the active power drawn by STATCOM, the modulation index can be maximized by adjusting DC link current. Within the scope of this thesis, an adaptive control of DC link current will be designed and applied to a 0.4kV 50kVAr three phase current source converter based STATCOM so that the power losses can be reduced. The theoretical work will be compared and discussed with the experimental results.

This research work is devoted to the analysis, design and development of the first medium power Current-Source Converter (CSC) based distribution-type Static Synchronous Compensator (D-STATCOM) with simplest converter topology and coupling transformer connection. The developed CSC-D-STATCOM includes a +/-750kVAr full-bridge CSC employing Selective Harmonic Elimination Method (SHEM), a 250kVAr low-pass input filter at 1kV voltage level, and a &amp / #916 / /Y connected coupling transformer for connection to medium-voltage load bus. The power stage of CSC is composed of series connection of natural air-cooled high-voltage IGCT switched at 500 Hz for the elimination of four lowest current harmonic components (5th, 7th, 11th, 13th), and optimized fast recovery high voltage diode in each leg. Reactive power control is achieved by applying the phase shift angle control at fixed modulation index, which is implemented digitally on a DSP microcontroller. The developed system has been implemented for compensation of rapidly varying reactive power demand of coal mining excavators in Turkish Coal Enterprises. The field test results have shown that the proposed CSC D-STATCOM serves as a technologically new full substitute of conventional Voltage-Source Converter based D-STATCOM having complex transformer connections in view of relatively fast response in reactive power compensation, very low total demand distortion factors, complying with the IEEE Std. 519-1992 even for the weakest power systems, and acceptable efficacy figures.

This research work is devoted to the analysis, design and development of the Current-Source Converter (CSC) based distribution-type Static Synchronous Compensator (D-STATCOM) for low-voltage applications in reactive-power control in order to achieve i) faster transient response in reactive-power control, ii) lower current harmonic distortion, iii) lower power losses and iv) minimum storage elements in comparison with conventional solutions. The developed CSC-D-STATCOM includes a low-pass input filter and a three phase forced-commutated CSC which is composed of six insulated gate bipolar transistors (IGBT) with built-in series diodes. The analysis and the control of the CSC-D-STATCOM are carried out in dq-synchronous reference frame in order to obtain the reference current waveform which is to be generated by switching the IGBTs at 3kHz with the use of space vector modulation.

A pulse hysteresisgraph system was constructed capable outputting current source and voltages source waveforms. MATLAB scripts were created to analyze the collected data. Three toroidal samples of soft magnetic materials were prepared. Theoretical modeling was done to predict the variation of effective applied magnetic fields inside the toroids from ideal assumptions due to three effects: wire spacing, cylindrical spreading, and eddy current generated fields. Data was collected under sinusoidal voltage source and sinusoidal current source excitation at 1 kHz. Large differences in core loss were noted especially at higher field excitations. Core loss under sinusoidal current source excitation was found to always be greater than or equal to core loss under sinusoidal voltage source. Normal magnetization curves under sinusoidal current and voltage source excitation were also compared. Significant differences were apparent in the magnetization curves of one sample toroid, and slight differences noted in the curves of the other two samples. Eddy currents were offered as a primary mechanism for the difference in core loss between sinusoidal current source and sinusoidal voltage source. A formula to predict the relative eddy current losses to be expected from an arbitrary, periodic voltage waveform shape is given. / text