Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-04-11T15:15:34Z No. of bitstreams: 1 joaotitoalmeidavianna.pdf: 4307092 bytes, checksum: 061d2a47a7e37bf6c1a1875c27f352ad (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-04-18T12:47:46Z (GMT) No. of bitstreams: 1 joaotitoalmeidavianna.pdf: 4307092 bytes, checksum: 061d2a47a7e37bf6c1a1875c27f352ad (MD5) / Made available in DSpace on 2017-04-18T12:47:46Z (GMT). No. of bitstreams: 1 joaotitoalmeidavianna.pdf: 4307092 bytes, checksum: 061d2a47a7e37bf6c1a1875c27f352ad (MD5) Previous issue date: 2016-09-05 / O presente trabalho tem por objetivo o desenvolvimento de alternativas para detecção de faltas de alta impedância (HIFs) sustentadas em sistemas de distribuição. Como a maioria destas faltas é monofásica, enfoque foi dado para este tipo de falta no desenvolvimento do trabalho. Devido ao desequilíbrio e a variação da carga dos sistemas de distribuição, a proteção tradicional dos mesmos apresenta dificuldades na detecção destas faltas, que acabam se sustentando no sistema. Isso acarreta um risco de segurança pública, na medida em que estruturas externas ao sistema elétrico são mantidas energizadas representando risco a vidas que entrem em contato com elas. Características da topologia típica de sistemas de distribuição brasileiros (conhecida como do “tipo europeu”) foram exploradas de forma a se elaborar duas propostas de métodos de detecção e localização de HIFs. Ambos os métodos se baseiam na alocação de medidores alocados ao longo do sistema de distribuição, cujas medidas são integradas a uma central de dados via rede de comunicação. Nesta central, as medidas recebidas são analisadas de forma a diagnosticar a presença e localização de faltas monofásicas de alta impedância no sistema e emitir alarmes que descrevam a situação atual do mesmo. O primeiro método, o PDSZ, baseia-se em medições fasoriais sincronizadas e, além da detecção e localização, conta com um algoritmo de classificação das fases envolvidas na falta. Já o segundo (PQDSZ) baseia-se em medições não fasoriais e não possui esta última funcionalidade. Ambos os métodos são implementados em laboratório e simulados com o uso de um Simulador Digital de Tempo Real (RTDS), o que permite uma validação bem próxima da aplicação dos métodos em um sistema real. Tais simulações comprovam a aplicabilidade dos métodos propostos e todos os resultados são analisados de forma a confirmar este fato. / This work aims to develop alternatives for the detection of sustained high impedance faults (HIFs) on distribution power systems. As the majority of these faults are single phase ones, the development was focused on this kind of fault. Due to load unbalance and variation of the distribution systems, the traditional protectiontechniques presents problems on detecting these faults, which are kept sustained on the system. This situation represents a serious public hazard, because the external electrical system structure remains energized, representing a risk to lives that eventually get in touch with it. There is also the risk of starting a fire from this fault point. Topologycal features of Brasilian distribution systems (known as “european type”) were exploited in order to elaborate two methods for detection and localization of HIFs. Both methods are based on the use of meters alocated along the distribution system, whose measurements are sent to a central data concentrator, through a network communication. Then, the received measurements are analysed in order to diagnose the ocurrence of single phase high impedance faults on the system and send alarms that describe the current system situation. The first method, named PDSZ, is based on Phasor Measurement Units (PMUs) and besides detecting and locating the fault, can also classify it, showing which phaseis involved. The second one, the PQDSZ, is based on non-synchronized measurements and can not classify the fault. Both methods were implemented on laboratory and tested using real equipament and a Real Time Digital Simulator (RTDS), which allows a very precise evaluation of the proposed methods, emulating conditions very similar the real ones. The simulations prove the aplicability of the proposed methods and the results are analysed in order to show the effectiveness of the proposed methods.

Reliability requirements for power systems have been constantly increasing, as customers of electric power desire high power availability. In order to improve continuity of supply in medium voltage (MV) networks, utilities in many countries in Europe, North/South America and Oceania use resonant earthing and allow the operation of the network during a single phasetoearth fault for periods from several seconds up to several hours, since the earthfault current at the fault point in such systems is considerably reduced and is unlikely to damage equipment or create hazardous voltages for people or livestock. Due to the neutral potential displacement during the phasetoearth fault, the healthy phases can rise up to 1.73 times the nominal values in the entire network, overstressing insulation and increasing the probability of a second fault from another phasetoearth fault somewhere in the network. The situation of two simultaneously active phasetoearth faults on different phases at different locations in a network is commonly called a crosscountry or doubleearth fault. The current through the earth then becomes many times higher than in case of a single phasetoearth fault in a resonantearthed network. Few studies about CrossCountry faults in MV systems have been done so far, particularly with real fault data and simulations. This thesis has as main objectives to: improve understanding of how crosscountry faults behave in resonantearthed systems; show how the double fault situation can be analytically calculated; and study what difficulties these faults can pose to traditional distance protection. This is done by: analyzing recordings of real faults, obtained from utilities in Scandinavia; modelling the fault situations analytically; simulating such systems with real system data; and examining in particular how faultedphase selection and distance protection behave during these faults. The developed mathematical model was compared with a simple simulated system. The results showed that the modelling produces accurate results. Analysis of the real faults showed that crosscountry faults cannot be equated directly to more traditional singlephase or doublephase faults on particular feeders, since voltages and currents can behave in atypical manners. Finally, during the comprehensive simulations of such network, limitations of traditional distance protection were identified. Phasetophase loops can face difficulties in properly detecting crosscountry faults even when one of the two simultaneous faults is relatively close to the measuring point. Phasetoearth loops can only be reliable when faults are solid or with very low resistance and in conditions where zero sequence current is present, while for multiple infeeds can also face difficulties. / Kraven på elkraftsystemets pålitlighet ökar hela tiden, då kundernas krav på tillgänglighet hela tiden stiger. Som ett led i att förbättra leveranssäkerheten på mellanspänningsnivån har ett flertal elnätsbolag i Europa, Nord och Sydamerika samt Oceanien börjat använda spoljordning. Genom detta kan jordfel under drift tillåtas, på allt ifrån sekundbasis till timmar. På detta sätt undviks driftavbrott vid jordfel. Spoljordning gör att jordfelsströmmen reduceras vilket minimerar risken för person, djur och egendomsskador. Under ett enfasigt jordfel höjs spänningen i de två opåverkade faserna till maximalt 1.73 gånger den nominella spänningen, detta i hela det aktuella nätet. Spänningshöjningen riskerar att ge skador på isolaringen, vilket gör att risken för uppkomsten av ett andra jordfel någonstans i samma nätavsnitt ökar. Att två enfasiga jordfel på olika platser inträffar samtidigt i samma nät, kallas dubbelt jordfel, på engelska crosscountry fault. Strömmen genom jord i ett spoljordat nät blir i detta fallet mycket högre än för ett enfasigt jordfel. Än så länge har bara ett fåtal studier med verkliga data och simuleringar gällande dubbla jordfel på mellanspänningsnivå utförts. Detta examensarbete syftar till att: förbättra förståelsen för hur dubbla jordfel uppkommer i spoljordade nät; beskriva hur dubbla jordfel kan beräknas analytiskt; och visa på vilka svårigheter dubbla jordfel kan ge för distansskydd. Detta görs genom: att analysera verkliga fel uppkomna i Skandinavien; analytisk modellering av dessa situationer; simulering med data från verkliga elnät; undersökning av hur felbehäftade delar och dess distansskydd beter sig vid fel. Den utvecklade matematiska modelleringsmetoden är jämförd med ett enkelt system som var simulerat, och resultaten av jämförelsen visar att modellen ger ett överensstämmande och noggrant resultat. Analysen av verkliga fel visar att dubbla jordfel inte direkt kan likställas med traditionella enfas- eller tvåfasfel för ett och samma utgående fack, då spänning och strömmar kan ha olika beteende. Avslutningsvis kunde begränsningar hos traditionella distansskydd upptäckas, tack var en omfattande simulering av nyss nämnda elnät. Fas- fas slingan kan se svårigheter att detektera dubbla fel även när ett av dessa två är nära till mätpunkten. Fas- jord slingan är endast pålitlig när felresistansen är försumbar, eller i fall där felresistansen är låg, samt i fall när nollföljdsströmmen är närvarande. Flera felströmsinmatningar kan likaså ställa till svårigheter.

This thesis introduces a new modeling approach that allows very large power systems to be modeled on a real time electro-magnetic transients (EMT) digital simulator with reduced hardware costs. The key step in achieving this is the development of an improved wide-band multi-port equivalent, which reduces a large power network into a small manageable equivalent model that preserves wideband behaviors. This approach has a foundation method that use a two part equivalent in which the high frequency behavior of the equivalenced network is represented by a terminating frequency dependent network equivalent (FDNE), with the low frequency behavior being modeled using a detailed Transient Stability Analysis (TSA) model that only models the electromechanical behavior. This approach allowed the modelling of medium size electric regions up to hundreds of buses in real time. This thesis extends the equivalent by implementing a reduced order of the detailed electromechanical TSA equivalent mentioned above. Coherency based reduction is used for the electromechanical model of the power network to be equivalenced, and is implemented as a Transient Stability Analysis (TSA) type electromechanical equivalent. A challenge in implementing the FDNE is to ensure that it is a passive network, as otherwise its inclusion could lead to unstable simulation. This thesis also introduces a practical procedure to enforce passivity in the FDNE. The validity of the proposed technique is demonstrated by comparing the approach with detailed electromagnetic simulations of the well-known 39 bus New England system and a modified 39 bus system with an HVDC infeed with coupling between the dc line and an adjacent ac line, in addition to a 108 bus ac system. The power of the method is demonstrated by the real-time simulation of a large system with 2300 busses and 139 generators. It has been shown that this approach has the potential to increase by at least one order of magnitude the size of the network that can be modeled and thus on a real time electro-magnetic transients (EMT) digital simulator with reduced hardware costs.

This thesis introduces a new modeling approach that allows very large power systems to be modeled on a real time electro-magnetic transients (EMT) digital simulator with reduced hardware costs. The key step in achieving this is the development of an improved wide-band multi-port equivalent, which reduces a large power network into a small manageable equivalent model that preserves wideband behaviors. This approach has a foundation method that use a two part equivalent in which the high frequency behavior of the equivalenced network is represented by a terminating frequency dependent network equivalent (FDNE), with the low frequency behavior being modeled using a detailed Transient Stability Analysis (TSA) model that only models the electromechanical behavior. This approach allowed the modelling of medium size electric regions up to hundreds of buses in real time. This thesis extends the equivalent by implementing a reduced order of the detailed electromechanical TSA equivalent mentioned above. Coherency based reduction is used for the electromechanical model of the power network to be equivalenced, and is implemented as a Transient Stability Analysis (TSA) type electromechanical equivalent. A challenge in implementing the FDNE is to ensure that it is a passive network, as otherwise its inclusion could lead to unstable simulation. This thesis also introduces a practical procedure to enforce passivity in the FDNE. The validity of the proposed technique is demonstrated by comparing the approach with detailed electromagnetic simulations of the well-known 39 bus New England system and a modified 39 bus system with an HVDC infeed with coupling between the dc line and an adjacent ac line, in addition to a 108 bus ac system. The power of the method is demonstrated by the real-time simulation of a large system with 2300 busses and 139 generators. It has been shown that this approach has the potential to increase by at least one order of magnitude the size of the network that can be modeled and thus on a real time electro-magnetic transients (EMT) digital simulator with reduced hardware costs.

Thesis (M.Sc. (Statistics)) -- University of Limpopo, 2020 / Background: Death and economic losses due to road traffic accidents (RTA) are huge global public health and developmental problems and need urgent attention. Each year nearly 1.24 million people die and millions suffer various forms of disability as a result of road accidents. This puts road traffic injuries (RTIs) as the eighth leading cause of death globally and RTIs are set to become the fifth leading cause of death worldwide by the year 2030 unless urgent actions are taken. Aim: In this paper, we investigate factors that contribute to road traffic deaths (RTDs) in the Limpopo province of South Africa using models such as the generalized linear models (GLM) and zero inflated models. Methods: The study was based on retrospective data that comprised of reports of 18,029 road traffic accidents and 4,944 road traffic deaths over the years 2009 – 2015. Generalized linear modelling and zero-inflated models were used to identify factors and determine their relationships to RTDs. Results: The data was split into two categories: deaths that occurred during holidays and those that occurred during non-holiday periods. It was found that the following variables, namely, Monday, human actions, vehicle conditions and vehicle makes, were significant predictors of RTDs during holidays. On the other hand, during non-holiday periods, weekend, Tuesday, Wednesday, national road, provincial road, sedan, LDV, combi and bus were found to be significant predictors of road traffic deaths. Conclusion: GLM techniques, such as the standard Poisson regression model and the negative binomial (NB) model, did little to explain the zero excess, therefore, zero-inflated models, such as zero-inflated negative binomial (ZINB), were found to be useful in explaining excess zeros. Recommendation: The study recommends that the government should make more human power available during the festive seasons, such as the December holidays, and over weekends.

The development of interconnected power systems together with an increasing number of renewable non-synchronous power sources, create major challenges for the power system to meet the voltage stability and power quality requirements. One way to increase the voltage stability in a sustainable way is to locally implement a STATCOM. By enhancing grid voltage stability under varying network conditions, the active power transfer capability will increase. However, before a STATCOM can be deployed in the power system, the behavior of it needs tobe investigated for the specific network conditions at the point of interface. The thesis develops a software model in RSCAD of a STATCOM along with important control functions for real-time simulations in RTDS without hardware-in-the-loop. The model aims to be sufficient for representing the gross behavior of a STATCOM in real-time simulations in order to get a quick overview of the dynamic response of the system. The model’s overall performance is evaluated through simulations in RTDS. The results indicate that the main control functions are operating in a stable and sufficient way. Hence, the model can perform in different operation modes as well as handling unbalances that are introduced in the system without losing controllability. There is potential for improvements in order to obtain a model with a more sophisticated control system. The main area would be to introduce limiters and anti-windups at appropriate places as well as a fault-ride-through logic to ensure a safe and stable operation during disturbances.

The Master’s Thesis is focused to introduction MES systems and factory information systems. Describes main advantages of applying MES systems and describes concept of Six Big Losses, which characterise main losses in production. For quantification of losses are used indicators like OEE. Next part is simulation data of Filler Machine and producing via OPC Server. Main application made in software SIMATIC IT by Siemens is collecting, processing and archiving data to PPA database due to rules and setups. Presentation of data is powered by Data Report which provides overview of efficiency and main losses in production.