No description available.

No description available.

This thesis investigates in detail a multilevel cascaded STATCOM employing indirect voltage control and square-wave control in combination as control strategy. An analytical steady-state model of a multilevel converter in harmonic domain is developed that can be used, for example, for resonance studies within a multilevel converter, or/and in studying interactions between a multilevel STATCOM and the host ac grid (like harmonic resonance). A systematic method is developed for STATCOM system design and optimization of STATCOM system parameters. The focus lies on minimizing losses, minimizing voltage. Total Harmonic Distortion (THD) and minimizing dc voltage ripple. Analytical formulae are presented that can be used to calculate the best value of each STATCOM system parameters. A discrete and an analytical dynamic converter model of a multilevel converter are developed to enable dynamic and/or stability studies. For the discrete model, the operating modes of a single-cell are analysed in detail and emulated using signal generators and integrators. The analytical multilevel converter model is segmented into a dynamic and static part in order to represent accurately all internal feedback connections. A general approach is developed for dynamic modelling of a STATCOM system. The dynamic system model has modular structure, and the controller gains are selected by analyzing the root locus of the analytical model to give optimum responses. The model is very accurate in the sub-synchronous range, and it is adequate for most control design applications and practical stability issues below 100 Hz. The controller robustness is also studied where the analytical STATCOM system model is used to perform eigenvalue analysis and to design controllers. Two different advanced control techniques are investigated and compared against the conventional method of proportional integral (PI) feedback voltage control.

Der Einfluß eines Seherstroms auf die Ausbreitung langer barotroper Wellen wird untersucht. Der Antarktische Zirkumpolarstrom ist als eine zonale Scherströmung zu betrachten. Wir zeigen auf einer ß-Ebene, daß der Antarktische Zirkumpolarstrom für zonal sich ausbreitende Wellen als Wellenleiter wirkt. Für den linearen Fall des Scherstromes wird die analytische Lösung angegeben. Die gebundenen Wellen sind meridional modal strukturiert und bewegen sich mit einer Gruppengeschwindigkeit von rund 0.02ms·1 von West nach Ost (alle). / We study the influence of a shear current on barotropic long wave propagation. The Antarctic Circumpolar Current can be considered as a sheared zonal flow. We show tht on a ß-plane the Antarctic Circumpolar Current acts as a waveguide for zonally propagating waves. We give the analytical solution for the linear sheared case. The trapped waves are modal structured in the meridional direction and propagate from the West to the East (all) with a group velocity of nearly 0.02ms·1.

Ce manuscrit présente les résultats d’une analyse exhaustive des mécanismes physiques qui limitent les performances et la fiabilité des transistors à haute mobilité d’électrons (HEMT) sur nitrure de gallium (GaN). En particulier : • Les phénomènes de dégradation à fort champ électrique des HEMT sur GaN sont analysés en comparant les données expérimentales avec les résultats de simulations physiques. Des stresses DC de 150 heures ont été effectués en conditions de canal ouvert et de pincement. Les effets des dégradations qui ont caractérisé ces deux types de stresses sont les suivants: une chute de courant DC de drain, une amplification des effets de gate-lag, et une diminution du courant inverse de grille. Les simulations physiques indiquent que la génération simultanée de piéges de surface (et/ou barrière) et de volume peut expliquer tous les modes de dégradation décrits plus haut. Les mesures expérimentales ont également montré que le stress en canal ouvert a causé une chute de la transconductance seulement pour de fortes valeurs de la tension VGS, alors que le stress au pincement a provoqué une chute de transconductance uniforme pour toutes les valeurs de VGS. Ce comportement peut être reproduit par la simulation physique pourvu que, dans le cas de stress a canal ouvert, on considère que les piéges s’accumulent au long d’une vaste région qui s’étend latéralement du bord de la grille vers le contact de drain, tandis que, dans le cas du stress au pincement, on considère que la génération des pièges ait lieu dans une portion plus petite de la zone d’accès à proximité de la grille et qu’elle soit accompagnée par une grande dégradation des paramètres de transport du canal. Enfin on propose que les électrons chauds et l’augmentation de la contrainte par le champ électrique soient à l’origine des dégradations observées après les stresses a canal ouvert et au pincement respectivement. • Les piéges dans les HEMT sur GaN ont été caractérisés en utilisant les techniques de DLTS et leur comportement associé de charge/décharge est interprété à l’aide des simulations physiques. Sous certaines conditions de polarisation, les piéges du buffer peuvent produire de faux signaux de piéges de surface, c'est-à-dire, le même type de signaux I-DLTS et ICTS attribués généralement aux piéges de surface. Clarifier cet aspect est très important à la fois pour les tests de fiabilité et pour l’optimisation des dispositifs, car il peut provoquer une identification erronée du mécanisme de dégradation, et par conséquent induire une mauvaise correction des procédés technologiques. • Les mécanismes physiques qui provoquent l’effondrement du courant RF dans les HEMT sur GaN sont analysés par le biais de mesures expérimentales et de simulations physiques. Ce travail propose les conditions suivantes : i) les piéges du buffer aussi bien que ceux de surface peuvent contribuer à l’effondrement du courant RF à travers un mécanisme identique qui impliquerait la capture et l’émission des électrons provenant de la grille; ii) la passivation de la surface diminue considérablement l’effondrement du courant RF par la réduction du champ électrique en surface et la diminution qui en découle de l’injection d’ électrons de la grille vers les pièges ; iii) pour des densités de piéges de surface inférieures à 9 × 1012 cm-2 , des barrières de potentiel superficiels dans l’ordre de 1-2 eV peuvent coexister avec des piéges de surface ayant des énergies plus faibles et qui causent l’effondrement du courant RF caractérisé par des constantes de temps relativement courtes. • Les effets de l’effondrement du courant dans les HEMT sur GaN sont étudiés en utilisant les résultats de mesures expérimentales et de simulations physiques. D’après les mesures pulsées, les dispositifs employés montrent un gate-lag considérable et un drain-lag négligeable qui peuvent être attribués à la présence de piéges de surface et de buffer respectivement. / This thesis reports the results of an extensive analysis of the physical mechanisms that limit the performance and reliability of gallium nitride (GaN) based High Electron Mobility Transistors (HEMT). In particular: • High electric field degradation phenomena are investigated in GaN-capped AlGaN/GaN HEMTs by comparing experimental data with numerical device simulations. Under power- and OFF-state conditions, 150-h DC stresses were carried out. Degradation effects characterizing both stress experiments were as follows: a drop in the dc drain current, the amplification of gate-lag effects, and a decrease in the reverse gate leakage current. Numerical simulations indicate that the simultaneous generation of surface (and/or barrier) and buffer traps can account for all of the aforementioned degradation modes. Experiments also showed that the power-state stress induced a drop in the transconductance at high gate–source voltages only, whereas the OFF-state stress led to a uniform transconductance drop over the entire gate-source-voltage range. This behavior can be reproduced by simulations provided that, under the power-state stress, traps are assumed to accumulate over a wide region extending laterally from the gate edge toward the drain contact, whereas, under the OFF-state stress, trap generation is supposed to take place in a narrower portion of the drain-access region close to the gate edge and to be accompanied by a significant degradation of the channel transport parameters. Channel hot electrons and electric-field-induced strain-enhancement are finally suggested to play major roles in power-state and off-state degradation, respectively. • Traps are characterized in AlGaN-GaN HEMTs by means of DLTS techniques and the associated charge/discharge behavior is interpreted with the aid of numerical device simulations. Under specific bias conditions, buffer traps can produce ‘‘false’’ surface-trap signals, i.e. the same type of current-mode DLTS (I DLTS) or gate-lag signals that are generally attributed to surface traps. Clarifying this aspect is important for both reliability testing and device optimization, as it can lead to erroneous identification of the degradation mechanism, thus resulting in wrong correction actions on the technological process. • The physical mechanisms underlying RF current collapse effects in AlGaN-GaN high electron mobility transistors are studied by means of measurements and numerical device simulations. This work suggests the following conclusions: i) both surface and buffer traps can contribute to RF current collapse through a similar physical mechanism involving capture and emission of electrons tunneling from the gate; ii) surface passivation strongly mitigates RF current collapse by reducing the surface electric field and inhibiting electron injection into traps; iii) for surface-trap densities lower than 9 × 1012 cm-2, surface-potential barriers in the 1–2 eV range can coexist with surface traps having much a shallower energy and, therefore, inducing RF current-collapse effects characterized by relatively short time constants. • Current collapse effects are investigated in AlGaN/GaN HEMTs by means of measurements and numerical device simulations. According to pulsed measurements, the adopted devices exhibit a significant gate-lag and a negligible drain-lag ascribed to the presence of surface and buffer traps, respectively. Furthermore, illumination of the devices with two specific wavelengths can result in either a recovering of current collapse or a decrease in the gate current. On the other hand, numerical device simulations suggest that the kink effect can be explained by electron trapping into barrier traps and the subsequent electron emission after a critical electric-field value is reached. / Questa tesi riporta i risultati ottenuti da un’ampia analisi dei meccanismi fisici che limitano le prestazioni e l’affidabilità dei transistor ad alta mobilità di elettroni (HEMT) al nitruro di gallio (GaN). In particolare: • I fenomeni di degradazione ad alto campo elettrico nei GaN/AlGaN/GaN HEMT sono analizzati confrontando i dati sperimentali con i risultati delle simulazioni numeriche. Sono stati effettuati stress DC di 150 ore in condizioni di canale aperto e chiuso. Gli effetti di degradazione che hanno caratterizzato entrambi i tipi di stress sono i seguenti: una caduta nella corrente DC di drain, un’amplificazione degli effetti di gate lag, e una diminuzione della corrente inversa di gate. Le simulazioni numeriche indicano che la generazione simultanea di trappole in superficie (e/o barriera) e buffer può spiegare tutti i suddetti modi di degradazione. Le misure sperimentali hanno mostrato inoltre che lo stress a canale aperto ha causato una caduta della tranconduttanza solo ad alte tensioni VGS, mentre lo stress a canale chiuso ha provocato una caduta della transconduttanza uniforme a tutte le tensioni VGS. Questo comportamento può essere riprodotto con le simulazioni se, nel caso di stress a canale aperto, si assume che le trappole si accumulano lungo un’ampia regione che si estende lateralmente dal bordo di gate verso il contatto di drain, mentre, nel caso di stress a canale chiuso, si suppone che la generazione delle trappole abbia luogo in una più stretta porzione della zona di accesso vicino al bordo di gate e che sia accompagnata da una degradazione significativa dei parametri di trasporto del canale. Infine si propone che gli elettroni caldi del canale e l’aumento di strain indotto dal campo elettrico siano alla base delle degradazioni osservate dopo gli stress a canale aperto e chiuso rispettivamente. • Le trappole in AlGaN-GaN HEMTs sono caratterizzate usando le tecniche di DLTS e il relativo comportamento di carica/scarica é interpretato con l’aiuto delle simulazioni numeriche. Sotto particolari condizioni di polarizzazione, le trappole di buffer possono produrre falsi segnali da trappole di superficie, ossia lo stesso tipo di segnali I-DLTS e forma d’onda di gate lag attribuiti generalmente alle trappole di superficie. Chiarire questo aspetto è molto importante sia per le prove di affidabilità che per l’ottimizzazione dei dispositivi, in quanto può provocare una errata identificazione del meccanismo di degradazione, portando ad azioni correttive sbagliate nell’ottimizzazione del processo tecnologico. • I meccanismi fisici che originano il collasso di corrente RF negli HEMT AlGaN-GaN sono analizzati usando misure sperimentali e simulazioni numeriche. Questo lavoro suggerisce le seguenti condizioni: i) sia le trappole di superficie che quelle di buffer possono contribuire al collasso di corrente RF tramite un simile meccanismo fisico che coinvolge la cattura e l’emissione di elettroni provenienti dal gate; ii) la passivazione della superficie diminuisce fortemente il collasso della corrente RF tramite la riduzione del campo elettrico in superficie e la conseguente diminuzione dell’iniezione di elettroni dal gate alle trappole; iii) per densità di trappole di superficie minori di 9 × 1012 cm-2 , barriere di potenziale superficiale di 1-2 eV possono coesistere con trappole di superficie aventi energie relativamente basse e che provocano effetti di collasso di corrente RF caratterizzati da costanti di tempo relativamente corte. • Gli effetti di collasso di corrente negli HEMT AlGaN-GaN sono studiati usando i risultati delle misure sperimentali e delle simulazioni numeriche. Basandosi sulle misure delle caratteristiche d’uscita impulsate, i dispositivi utilizzati mostrano un evidente gate-lag e un trascurabile drain-lag, attribuiti alla presenza di trappole di superficie e buffer rispettivamente.

Historically the nutrient rich Benguela ecosystem supported large stocks of commercially important fish which sustained the Namibian fishing sector. Recently, non-exploited species i.e. bearded goby (Sufflogobius bibarbatus) and jellyfish (Chrysaora fulgida and Aequorea forskalea) have become more apparent and are described as key-species in this ecosystem. Empirical evidence for understanding the stock abundance and dynamics of S. bibarbatus and jellyfish are still sparse, as research focus has been mainly on commercially important fish. The abundance of these non-exploited species in relation to the environment and commercial species are also not well understood. Lack of methods to effectively assess jellyfish and S. bibarbatus have furthermore limited our knowledge. Acoustics can cover large water volumes and observe many trophic groups and interactions simultaneously hence currently proposed as the most reliable observation tool available to remotely study multiple species that are overlapping and widely distributed in marine ecosystems. For acoustic assessments using echo sounders, the ability to detect, identify and distinguish targets from each other and the echo ability (target strength: TS) of individual targets is pivotal to convert acoustic data from a calibrated system into significant biological measures. The lack of effective acoustic identification (ID) techniques and knowledge about TS of species may limit the application of acoustics. The swimbladder generally contributes more than 90% to the backscattered energy from fish, which makes knowledge of the swimbladder vital for understanding the acoustic properties of a fish. Prior to this study, the presence or absence of a swimbladder within S. bibarbatus has been uncertain. This thesis is an exploratory study addressing 1) the acoustic identification challenge of species in aggregating in mixed assemblages and 2) the acoustic characteristics of the target species. The latter two are of essence to assess the biomass, distributions and ecological interactions of these non-exploited. The multiple frequency data (18, 38, 70, 120 and 200 kHz) and trawl data used in this study were collected on a survey conducted by the RV G.O.Sars during April 2008 in the northern Benguela. Fifteen validated assumed to be ‘single species' trawl and acoustic datasets were selected and used in the application and developing of ID techniques. Traditional acoustic identification techniques (Sᵥ-differencing and relative frequency response r(f)) were adopted and found ineffective as standalones to discriminate the species under study. The overlaps in the Sᵥ differences of the three species complicated separation. A multivariate statistical approach, Linear Discriminant Analysis (LDA) was applied to predict which of the variables s[subscript(A)], S[subscript(A)], Δs[subscript(A)] and r(f) discriminated the three species groups from each other with a higher accuracy. It was found that by combining backscattering strength S[subscript(A)] and r(f) a correct classification accuracy of up to 95% could be obtained. Limitation is that the LDA technique as any classification method is not applicable in “real time” during surveys. A new technique, here within referred to as the Separator Technique, which incorporates the standard techniques, LDA results, a novel r(f) similarity comparison technique and a threshold s[subscript(A)] response technique was established. The effectiveness of the Separator Technique is in the recognition of similarities and stability in frequency response by simple correlation of the observed frequency response at systematic Sᵥ-threshold levels. Accurate acoustic classification depends on good and valid training datasets and there has so far not been a simple way of acoustically detecting if the selected assumed “pure” datasets is contaminated or not. Only available reliable source are the trawl samples. The r(f) similarity comparison method showed that some of the assumed ‘single species' trawls were mixed and that presence of <1% of strong scatterers could mask a weaker scatterer. By evaluating the threshold s[subscript(A)] frequency response, the proportion of thresholded backscattering could be quantified. A frequency which is more appropriate for the acoustic assessment of the respective species in mixed aggregations could also be identified. Further improvements of the Separator Technique are required in terms of the precise Sᵥ-cut levels. The presence of S. bibarbatus' swimbladder was confirmed from two thawed specimens. From further investigations on 26 dissections of sampled S. bibarbatus, the swimbladder was identified as a physoclist (closed swimbladder) with an extensive gas gland, and its morphology was roughly described as prolate spheroid shaped and with about 5ᴼ negatively tilted compared fish vertebra. This means that the strongest echo from a goby will be found when the fish is at about 5ᴼ head down relative to the horizontal. The in situ TS of 8 cm sized S. bibarbatus and the two jellyfish species: C. fulgida [umbrella diameter: 21.7 cm] and A. forskalea [16 cm] at multiple frequencies (18, 38, 70, 120 and 200 kHz) was estimated. At 38 kHz, the TS was -53 dB for S. bibarbatus, -58 dB for A. forskalea and -66 dB for C. fulgida. The single echo detection (SED) approach which is assumed to be a more accurate method for estimating TS than the previously applied methods for jellyfish. The TS results for S. bibarbatus are of similar magnitude to other published TS values of C. fulgida. This suggests that estimates of jellyfish may be overestimated due to inaccuracies in target identification. This thesis established the acoustic characteristics of jellyfish and S. bibarbatus within the northern Benguela which makes it possible to acoustically assess and monitor jellyfish and/or fish. The identification technique though still in early phases of development, can be applied to enhance quality of training datasets (samples) used in classification. This piece of work can reduce variability in biomass estimates that arises from masking or misclassification of echoes.

M.Ing. / The drive for smaller and higher density power supplies have been realised by advances in switching technologies, higher frequencies and smaller components. Along with the advances of higher switching frequencies, came a number of restrictive parasitic effects that were insignificant at lower frequencies (in use a few years ago). A problem that is becoming of increasing concern, as the frequencies increase, rise times decrease and current levels increase, is the reactance of the parasitic inductance in voltage fed converter. This inductance is responsible for a multitude of limitations and problems in high frequency converters, with the most important being unstable voltage supplies, large voltage spikes during switching (which leads to electromagnetic interference), and power transfer limitations. The main contributors of this parasitic inductance was found to be the inherent inductance of the conductors of the DC bus, the internal inductance of the capacitor elements used in the DC bus and the paralleling of these capacitor elements (capacitor bank). It was decided to investigate the cause of these identified inductances in an attempt at finding a means to reduce them, thereby improving the performance of the converter. This was accompanied by a search into prediction methods for the inductance and capacitance of the DC bus conductors. The ability to predict the inductance and capacitance inherent to the DC bus conductors, will allow for a large decrease in prototyping, and should give insight into the causes of these elements and how to manipulate them. This was done for the DC bus conductors, and led to insight into their inductance and capacitance origins. Means to reduce this inductance was found, along with the ability to predict the inductance and capacitance of a number of DC bus conductors. The last two identified parasitic inductance sources, the internal inductance of the capacitors and inductance of the capacitor bank, were then investigated. The cause of the inductance in the capacitor elements was discovered, along with the factors on which the capacitor elements are dependent. A great deal of the inductance, and its associated effects, can be avoided through proper capacitor selection and correct capacitor bank design. In order to bring this study in context with a practical scenario, the information previously obtained was incorporated in a full bridge voltage fed converter. The previous findings on inductance and capacitance held equally well when applied to a practical scenario. Additional means to reduce the effects of the parasitic inductances were discovered, and the inductance and capacitance prediction methods proved to be relatively accurate when applied to the DC bus conductors of a physical converter.

ABSTRACT Power transformers are key components for electrical energy transfer in a power system. Stability and security of transformer protection are important to system operation. Many maloperation cases of transformer differential protection are caused by inrush current problems. The phenomenon of transformer inrush current has been widely discussed in literature. Therefore, this research only discussed and analyzed inrush current problems to transformer differential protection. To understand the inrush current problems on differential protection, transformer simulation models were presented using Matlab/Simulink. Differential relay simulations for internal and external faults, and relay performance during current transformer saturation were performed using PSCAD. Recommendations were made on methods of reducing the blocking time of differential protection during inrush conditions.

Thesis is submitted in fulfilment of the requirements for the degree of Master of Science in Engineering to the Faculty of Engineering and Built Environment, University of the Witwatersrand, Johannesburg 2018 / Power electronic converters that convert DC to AC, or vice versa, require an energy buffer between the AC and DC ports of the converter to compensate for the instantaneous power mismatch. Electrolytic capacitors are mostly used for these buffering applications because of the high energy density when compared to other capacitors, but unfortunately this type of capacitor also has low reliability. This dissertation proposes a general solution from a fundamental approach to solve the required capacitor power requirements on the DC-bus of an inverter. From the resulting model, an alternative active filter design technique to reduce the required capacitance of the DC-bus capacitor of a single phase inverter is presented. In this model, the minimum and maximum voltages of the capacitor can be chosen and the corresponding waveforms are calculated. An optimum region for the choice of capacitor voltage is shown to visually illustrate the trade-offs between the capacitor voltage, capacitance and converter losses. In this optimum area the reduction in capacitance is enough to allow the elimination of electrolytic capacitors, while maintaining comparable volume. In this technique, the DC-bus capacitor is decoupled from the DC-bus to allow wide voltage variation and the power processed by the capacitor is directly controlled, instead of the bus voltage. The allowable voltage variation of the capacitor can also be selected to fit the application or traded off in favour of capacitance as chosen by the designer. This general solution is applicable to any bi-directional converter used to decouple the capacitor from the DC-Bus / XL2018

Faculty of Engineering & Built Environment, School of Electrical & Info Engineering, Dissertation / The primary objective of this paper is to present results regarding data obtained from Eskom’s Lightning Positioning and Tracking System (LPATS) and is a continuation of the work presented at the two SAUPEC Conferences in Pretoria and Stellenbosch [1, 2]. LPATS provides some useful information regarding the lightning field measurements around the Brixton and Hillbrow Towers, in Johannesburg, for the two seasons of June 2001 to June 2003. The results suggest that there is a significant increase in apparent ground flash density in the vicinity of the towers when compared to the surrounding areas. The observation of mean current values in the order of -20kA suggests that the increased contribution of upward flashes to the total incidence of flashes in tall structures should lead to a decrease in measured current amplitudes.