In the fast growing field of Telecommunications, the back up DC power supply plays a vital role in powering the telecom equipment. This DC power supply is a combination of AC-DC Rectifier coupled with a battery bank to support the load when AC input is not available. Figures 0.1 and 0.2 show the line diagram of the DC power supply. The power supply is the most critical element in a telecom installation and it should be highly reliable in order to have un-interrupted service. (Fig) Besides reliability, power density and cost are the driving forces behind the success of a power supply in the market. Off late, the reach of telecom in the society is very wide covering remote villages and major metros. Given this environment, the power supply is exposed to extreme input conditions. It is desirable to design the power supply capable of withstanding wide AC input conditions. Another advantage is that the rectifier unit will keep the battery charged so that the battery will have long life. This thesis is aimed at designing a 1400W (56V/25A) telecom power supply, keeping in view of the issues expressed above. The aim is to design a Switched Mode Rectifier (SMR) that tolerate wide input voltage variations (90Vac to 300Vac). In addition, the design covers unity input power factor, high efficiency (> 90%), high power density ( ), parallel operation and low cost ( ). Chapter 1 of this thesis covers the context and motivation of the work. Chapter 2 presents the design issues pertaining to power supplies. The normalized description of the power converters is presented. Such a description enables one to compare several circuit topologies in order to make effective design decisions. In a similar way the effectiveness of the switches and mgnetics are presented to enable design decisions in the output stage of the rectifier. Chapter 3 presents the design of the 1400W telecom power supply, keeping in view of the stated specifications. The performance results of the converter are presented in Chapter 4. All the design goals have been met. The design exercise has also given insights into possible further improvements. Contributions from this work and course of future development work are indicated in the concluding chapter.

We study the problem of joint congestion control, routing and MAC layer scheduling in multi-hop wireless mesh networks, where the nodes in the network are subjected to energy expenditure rate constraints. As wireless scenario does not allow all the links to be active all the time, only a subset of given links can be active simultaneously. We model the inter-link interference using the link contention graph. All the nodes in the network are power-constrained and we model this constraint using energy expenditure rate matrix. Then we formulate the problem as a network utility maximization (NUM) problem. We notice that this is a convex optimization problem with affine constraints. We apply duality theory and decompose the problem into two sub-problems namely, network layer congestion control and routing problem, and MAC layer scheduling problem. The source adjusts its rate based on the cost of the least cost path to the destination where the cost of the path includes not only the prices of the links in it but also the prices associated with the nodes on the path. The MAC layer scheduling of the links is carried out based on the prices of the links. The optimal scheduler selects that set of non-interfering links, for which the sum of link prices is maximum. We study the effects of energy expenditure rate constraints of the nodes on the maximum possible network utility. It turns out that the dominant of the two constraints namely, the link capacity constraint and the node energy expenditure rate constraint affects the network utility most. Also we notice the fact that the energy expenditure rate constraints do not affect the nature of optimal link scheduling problem. Following this fact, we study the problem of distributed link scheduling. Optimal scheduling requires selecting independent set of maximum aggregate price, but this problem is known to be NP-hard. We first show that as long as scheduling policy selects the set of non-interfering links, it can not go unboundedly away from the optimal solution of network utility maximization problem. Then we proceed and evaluate a simple greedy scheduling algorithm. Analytical bounds on performance are provided and simulations indicate that the greedy heuristic performs well in practice.

The main objectives of this research are (a) to develop advanced simulation methods for voltage-recovery phenomena using improved, realistic system models and accurate solution techniques and (b) to develop methods for the mitigation of problems related to slow voltage recovery. Therefore, this work concentrates on the areas of voltage-recovery analysis in electric power systems, dynamic load modeling with emphasis on induction-motor models, dynamic simulation with emphasis on the numerical integration methods, and optimal allocation and operation of static and dynamic VAr resources. In the first part of this work, a general framework for power-system analysis is presented the main characteristics of which are (a) the utilization of full three-phase models and (b) the use of a "quadratized" mathematical formulation, which models the system under study as a set of mathematical equations of order no more than two. The modeling approach is essentially the same for steady-state, quasi-steady-state, and dynamic analysis. Furthermore, a new approach for time-domain transient simulation of electric power systems and dynamical systems, in general, is introduced in this research. The new methodology has been named quadratic integration method. The method is based on a numerical integration scheme that assumes that the system states vary quadraticaly within an integration time step. Accurate modeling and simulation of voltage-recovery phenomena allows the development of mitigation methodologies via the optimal allocation and operation of static and dynamic VAr resources over the planning horizon. This problem is solved with successive dynamic programming techniques with the following two innovations: (a) the states at each stage (candidate solutions) are obtained with static and dynamic (trajectory) sensitivity analysis and (b) each candidate solution is evaluated by considering the optimal operation of installed static and dynamic VAr sources utilizing concepts from the theory of applied optimal control and trajectory optimization.

Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro / O objetivo deste trabalho é conhecer e compreender melhor os imprevistos no fornecimento de energia elétrica, quando ocorrem as variações de tensão de curta duração (VTCD). O banco de dados necessário para os diagnósticos das faltas foi obtido através de simulações de um modelo de alimentador radial através do software PSCAD/EMTDC. Este trabalho utiliza um Phase-Locked Loop (PLL) com o intuito de detectar VTCDs e realizar a estimativa automática da frequência, do ângulo de fase e da amplitude das tensões e correntes da rede elétrica. Nesta pesquisa, desenvolveram-se duas redes neurais artificiais: uma para identificar e outra para localizar as VTCDs ocorridas no sistema de distribuição de energia elétrica. A técnica aqui proposta aplica-se a alimentadores trifásicos com cargas desequilibradas, que podem possuir ramais laterais trifásicos, bifásicos e monofásicos. No desenvolvimento da mesma, considera-se que há disponibilidade de medições de tensões e correntes no nó inicial do alimentador e também em alguns pontos esparsos ao longo do alimentador de distribuição. Os desempenhos das arquiteturas das redes neurais foram satisfatórios e demonstram a viabilidade das RNAs na obtenção das generalizações que habilitam o sistema para realizar a classificação de curtos-circuitos. / The objective of this work is to know and understand the unforeseen in the supply of electricity, when there are short duration voltage variations (SDVV). The required databases for the diagnosis of faults were obtained through simulations of a model of radial feeder through software PSCAD/EMTDC. This work uses a Phase-Locked Loop (PLL) in order to detect and perform the estimation SDVV automatic frequency, phase angle and amplitude of the voltage and current from the power grid. This research is developing two artificial neural networks: one to identify and another to locate the SDVV occurred in the distribution system of electricity. The technique proposed here applies to three-phase feeders with unbalanced loads, which can have side extensions triphasic, biphasic and monophasic. In developing the same, it is considered that there is availability of measurements of voltages and currents at the node of the initial feeder and also in some points scattered along the distribution feeder. The performances of the architectures of neural networks were satisfactory and demonstrate the feasibility of ANNs in obtaining the generalizations that enables the system for the classification of short circuits.

This dissertation deals with the challenge of power utility in South Africa which is on proactive detection of fallen power line conductors and real time sagging measurement together with slipping of such conductors. Various methods which are currently used for sag detection were characterized and evaluated to the aim of the research. A mathematical reconstruction done to estimate the lowest point of the conductor in a span is presented. Practical simulations and application of radio frequency identification (RFID) for sag detection is attempted through matlab software. RFID radar system is then analyzed in different modes and found to give precision measurement for sag in real time as opposed to global positioning system (GPS) if one dimension of the tag assumed fixed on the power line. Lastly errors detected on the measurements are corrected using a trainable artificial neural network. A conclusion is made by making recommendations in the advancement of the research. / Electrical Engineering / M. Tech. (Electrical Engineering)

Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro / O objetivo deste trabalho é conhecer e compreender melhor os imprevistos no fornecimento de energia elétrica, quando ocorrem as variações de tensão de curta duração (VTCD). O banco de dados necessário para os diagnósticos das faltas foi obtido através de simulações de um modelo de alimentador radial através do software PSCAD/EMTDC. Este trabalho utiliza um Phase-Locked Loop (PLL) com o intuito de detectar VTCDs e realizar a estimativa automática da frequência, do ângulo de fase e da amplitude das tensões e correntes da rede elétrica. Nesta pesquisa, desenvolveram-se duas redes neurais artificiais: uma para identificar e outra para localizar as VTCDs ocorridas no sistema de distribuição de energia elétrica. A técnica aqui proposta aplica-se a alimentadores trifásicos com cargas desequilibradas, que podem possuir ramais laterais trifásicos, bifásicos e monofásicos. No desenvolvimento da mesma, considera-se que há disponibilidade de medições de tensões e correntes no nó inicial do alimentador e também em alguns pontos esparsos ao longo do alimentador de distribuição. Os desempenhos das arquiteturas das redes neurais foram satisfatórios e demonstram a viabilidade das RNAs na obtenção das generalizações que habilitam o sistema para realizar a classificação de curtos-circuitos. / The objective of this work is to know and understand the unforeseen in the supply of electricity, when there are short duration voltage variations (SDVV). The required databases for the diagnosis of faults were obtained through simulations of a model of radial feeder through software PSCAD/EMTDC. This work uses a Phase-Locked Loop (PLL) in order to detect and perform the estimation SDVV automatic frequency, phase angle and amplitude of the voltage and current from the power grid. This research is developing two artificial neural networks: one to identify and another to locate the SDVV occurred in the distribution system of electricity. The technique proposed here applies to three-phase feeders with unbalanced loads, which can have side extensions triphasic, biphasic and monophasic. In developing the same, it is considered that there is availability of measurements of voltages and currents at the node of the initial feeder and also in some points scattered along the distribution feeder. The performances of the architectures of neural networks were satisfactory and demonstrate the feasibility of ANNs in obtaining the generalizations that enables the system for the classification of short circuits.

Thesis (MTech (Electrical Engineering))--Cape Peninsula University of Technology, 2009 / The purpose of this project is to develop methods for distribution network power quality' variations monitoring. Power quality (PO) has become a significant issue for both power suppliers and customers. There have been important changes in power system regarding to power quality requirements. "Power quality" is the combination at voltage quality and current quality. The main research problem of the project is to investigate the power quality of a distribution network by selection of proper measurement, applying and developing the existing classic and modern signal conditioning methods for power disturbance's parameters extracting and monitoring. The research objectives are: To study the standard lEC 61000-4-30 requirements. to investigate the common couplings in the distribution network. To identity the points for measurement, to develop MySQL database for the data from the measurement and to develop MATLAB software tor simulation of the network To develop methods based on Fourier transforms for estimation of the parameters of the disturbances. To develop software for the methods implementation, The influence of different loads on power quality disturbances are considered in the distribution network. Points on the network and meters according to the lEC power quality standards are investigated and applied for the CPUT Bellville campus distribution network. The implementation of the power quality monitoring for the CPUT Bellville campus helps the quality of power supply to be improved and the used power to be reduced. MATLAB programs to communicate with the database and calculate the disturbances and power quality parameters are developed.

CAPES / Este trabalho apresenta novas abordagens para duas das etapas fundamentais relacionadas com análise automática de oscilografias de redes de distribuição: a detecção dos instantes transitórios e a sua classificação. Para comparação e validação dos métodos são utilizadas duas bases de dados, sendo uma delas formada por dados simulados no aplicativo Alternative Transient Program e outra contendo dados reais de oscilógrafos instalados em uma rede de distribuição de energia elétrica. Os dados reais apresentam um conjunto de eventos relevante para as análises aqui propostas, principalmente por conter uma gama variada de eventos, incluindo transitórios decorrentes de descargas atmosféricas. Com relação à detecção de transitórios (segmentação de oscilografias), foram testados os métodos atualmente propostos na literatura, os quais contemplam Filtro de Kalman, Transformada Wavelet Discreta e Modelos Autorregressivos, além de serem propostas duas novas técnicas baseadas no Operador de Energia de Teager e Representação de Dados Utilizando Vetores Suporte. Demonstra-se que, tanto para dados simulados quanto para dados reais, o método de detecção baseado na Representação de Dados utilizando Vetores Suporte aponta para um melhor desempenho global no processo de detecção. Com relação à classificação automática de oscilografias, propõe-se uma nova abordagem incluindo um estágio dedicado à detecção de padrões não inseridos no aprendizado prévio do classificador, denominados de novidades, além da própria classificação multiclasse normalmente empregada para diferenciar múltiplas classes conhecidas a priori. São testadas abordagens utilizando a detecção de novidades e classificação multiclasse em estágios simultâneos e subsequentes, com base nos classificadores X-Médias, K-Vizinhos-Mais-Próximos e Representação de Dados Utilizando Vetores Suporte com diferentes formulações, além do próprio classificador multiclasse baseado em Máquinas de Vetor Suporte. Adicionalmente, é proposto um tratamento aos padrões considerados como novidades, com o intuito de fornecer informações ao especialista sobre as similaridades existentes entre os padrões desse conjunto. Para realizar esse processo, optou-se por utilizar modelos de agrupamento automático. Os resultados finais, principalmente para a base de dados incluindo eventos reais, mostram que é possível obter um desempenho de classificação relevante (acima de 80%) para cada um dos estágios do processo de classificação proposto, o qual inclui a detecção de novidades, a classificação multiclasse e o processamento de padrões classificados como novidades (agrupamento automático). / This work presents new approaches for two of the fundamental steps in automatic waveform analysis in electrical distribution systems: transient time detection and its classification. Two datasets were used to compare and validate the proposed methods. The first is composed by simulated waveforms, by using the Alternative Transient Program, while the second is formed by real data from a monitoring system developed for overhead distribution power lines. The real data present a set of relevant events for the analysis proposed here, mainly due to the variety of events, including lightning-related transients. Regarding transient detection (waveform segmentation), the experiments involve usual segmentation methods, such as Kalman filtering, standard Discrete Wavelet Transform, and autoregressive models, besides two new techniques based on the Teager Energy Operator and Support Vector Data Description. The results obtained on both simulated and real world data demonstrate that the method based on Support Vector Data Description outperforms other methods in the transient identification task. Regarding the automatic waveform classification, a new approach including the detection of classes not defined in the training stage (called novelties) is presented. Also, the classifier is able to discriminate among multiple known classes, normally defined as multi-class classification. Two different approaches are compared, by using multi-class classification and novelty detection in two subsequent stages and in a simultaneous way. The following classifiers were assessed: X-Means, K-Nearest-Neighbors, and Support Vector Data Description with different formulations, besides the Support Vector Machine for multi-class classification. Furthermore, a technique for the post-processing of the novelties is presented, in order to provide some useful information for the experts, regarding possible similarities in the novelty set. To accomplish this task, automatic clustering methods were used. The final results, especially for the dataset with real examples, show that it is possible to obtain a relevant classification performance (above 80%) in each one of the three stages of the classification process: multi-class classification, novelty detection, and the post-processing applied to the novelties (automatic clustering).

CAPES / Este trabalho apresenta novas abordagens para duas das etapas fundamentais relacionadas com análise automática de oscilografias de redes de distribuição: a detecção dos instantes transitórios e a sua classificação. Para comparação e validação dos métodos são utilizadas duas bases de dados, sendo uma delas formada por dados simulados no aplicativo Alternative Transient Program e outra contendo dados reais de oscilógrafos instalados em uma rede de distribuição de energia elétrica. Os dados reais apresentam um conjunto de eventos relevante para as análises aqui propostas, principalmente por conter uma gama variada de eventos, incluindo transitórios decorrentes de descargas atmosféricas. Com relação à detecção de transitórios (segmentação de oscilografias), foram testados os métodos atualmente propostos na literatura, os quais contemplam Filtro de Kalman, Transformada Wavelet Discreta e Modelos Autorregressivos, além de serem propostas duas novas técnicas baseadas no Operador de Energia de Teager e Representação de Dados Utilizando Vetores Suporte. Demonstra-se que, tanto para dados simulados quanto para dados reais, o método de detecção baseado na Representação de Dados utilizando Vetores Suporte aponta para um melhor desempenho global no processo de detecção. Com relação à classificação automática de oscilografias, propõe-se uma nova abordagem incluindo um estágio dedicado à detecção de padrões não inseridos no aprendizado prévio do classificador, denominados de novidades, além da própria classificação multiclasse normalmente empregada para diferenciar múltiplas classes conhecidas a priori. São testadas abordagens utilizando a detecção de novidades e classificação multiclasse em estágios simultâneos e subsequentes, com base nos classificadores X-Médias, K-Vizinhos-Mais-Próximos e Representação de Dados Utilizando Vetores Suporte com diferentes formulações, além do próprio classificador multiclasse baseado em Máquinas de Vetor Suporte. Adicionalmente, é proposto um tratamento aos padrões considerados como novidades, com o intuito de fornecer informações ao especialista sobre as similaridades existentes entre os padrões desse conjunto. Para realizar esse processo, optou-se por utilizar modelos de agrupamento automático. Os resultados finais, principalmente para a base de dados incluindo eventos reais, mostram que é possível obter um desempenho de classificação relevante (acima de 80%) para cada um dos estágios do processo de classificação proposto, o qual inclui a detecção de novidades, a classificação multiclasse e o processamento de padrões classificados como novidades (agrupamento automático). / This work presents new approaches for two of the fundamental steps in automatic waveform analysis in electrical distribution systems: transient time detection and its classification. Two datasets were used to compare and validate the proposed methods. The first is composed by simulated waveforms, by using the Alternative Transient Program, while the second is formed by real data from a monitoring system developed for overhead distribution power lines. The real data present a set of relevant events for the analysis proposed here, mainly due to the variety of events, including lightning-related transients. Regarding transient detection (waveform segmentation), the experiments involve usual segmentation methods, such as Kalman filtering, standard Discrete Wavelet Transform, and autoregressive models, besides two new techniques based on the Teager Energy Operator and Support Vector Data Description. The results obtained on both simulated and real world data demonstrate that the method based on Support Vector Data Description outperforms other methods in the transient identification task. Regarding the automatic waveform classification, a new approach including the detection of classes not defined in the training stage (called novelties) is presented. Also, the classifier is able to discriminate among multiple known classes, normally defined as multi-class classification. Two different approaches are compared, by using multi-class classification and novelty detection in two subsequent stages and in a simultaneous way. The following classifiers were assessed: X-Means, K-Nearest-Neighbors, and Support Vector Data Description with different formulations, besides the Support Vector Machine for multi-class classification. Furthermore, a technique for the post-processing of the novelties is presented, in order to provide some useful information for the experts, regarding possible similarities in the novelty set. To accomplish this task, automatic clustering methods were used. The final results, especially for the dataset with real examples, show that it is possible to obtain a relevant classification performance (above 80%) in each one of the three stages of the classification process: multi-class classification, novelty detection, and the post-processing applied to the novelties (automatic clustering).

Submitted by Daniella Sodre (daniella.sodre@ufpe.br) on 2015-04-10T16:37:01Z No. of bitstreams: 2 TESE Rômulo Fernando Teixeira Vilela.pdf: 3159637 bytes, checksum: d8b68b1fd93d79fe6162c4abdd0b1aa0 (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) / Made available in DSpace on 2015-04-10T16:37:01Z (GMT). No. of bitstreams: 2 TESE Rômulo Fernando Teixeira Vilela.pdf: 3159637 bytes, checksum: d8b68b1fd93d79fe6162c4abdd0b1aa0 (MD5) license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Previous issue date: 2013-02-27 / Com os avanços em hardware, a engenharia de confiabilidade nos últimos 30 anos, tem nos mostrado equipamentos e sistemas complexos com níveis de falha muito baixos. Sistemas complexos na indústria nuclear, aeroespacial, química, elétrica entre outras possuem hoje em dia equipamentos e sistemas com níveis de confiabilidade que tem atendido adequadamente a sociedade. Entretanto, a operação e manutenção destes sistemas não dependem exclusivamente do desempenho intrínseco dos correspondentes equipamentos, dependem também da ação humana. Grandes acidentes no passado recente como Chernobyl, Bhopal, da nave Challenger e os grandes apagões no Brasil, colocaram em evidência a necessidade de redução do erro humano em sistemas complexos. A análise da confiabilidade humana surge assim como um apoio para a análise destes sistemas de operação e manutenção. Desde a década de 80 alguns avanços foram surgindo no estudo da confiabilidade humana. Técnicas como THERP, ATHEANA, CREAM e IDAC, se consolidaram ao longo do tempo como boas aplicações práticas para estudar, medir e prever o erro humano. Porém os fatores de desempenho utilizados em quase todas as técnicas supracitadas, tem se mostrado difíceis de serem estimados de um ponto de vista particular. Além disso, as particularidades do setor Hidroelétrico de Potência, definidas nos Procedimentos de Rede do Operador Nacional do Sistema (ONS) e nos instrumentos normativos da Agencia Reguladora ANEEL têm levado a necessidade de uma taxonomia que possa se adaptar a este importante e estratégico setor. Nesta tese, é proposta uma taxonomia e um modelo da sequência do processo de erro, para avaliação deste erro humano especificamente concebido para atender ao contexto de operação e manutencão do Sistema Hidroelétrico de Potência. Para ilustrar a nova taxonomia, foram coletados e analisados dados de cerca de dez anos de registro de erro humano de uma empresa de geração e transmissão de energia elétrica brasileira. Foram coletados 605 relatórios de desligamento por erro humano desde 1998 até 2009. Uma metodologia BBN-Base para a quantificação do erro humano é também discutida. A taxonomia e o modelo da sequência do processo de erro humano tanto quanto o modelo BBN-Based são ilustrados via um exemplo de uma aplicação no contexto de uma indústria Brasileira Hidroelétrica de Potência.-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------With advances in hardware reliability engineering in the last 30 years, we have seen equipment and complex systems with very low levels of failure. Complex systems in the nuclear industry, aerospatiale sector, chemical industries, electrical industries and others now have equipment and systems with levels of reliability that has adequately served the society. However, the operation and maintenance of these systems do not rely solely on intrinsec performance of the corresponding equipment, but they also depend on human action. Major accidents in the recent past such Chernobyl, Bhopal, the Challenger shuttle and major recent power blackouts in Brazil, highlighted the need to reduce human error in complex systems. The human reliability assessment emerges as a support to the analisys of the operation and maintenance of these type of systems. Since the late 80th some advances have emerged in the study of human reliability. Techniques such as THERP, ATHEANA, CREAM and IDAC, have been consolidated over time for the study, measure and prediction of human error. However performance shaped factors used in almost all the aforementioned techniques have proven difficult to be estimated from a practical standpoint. In addition, the specifics of the Hydroelectric Power Industry defined in the Grid Procedures of the National System Operator (Operador Nacional so Sistema, ONS) and the regulatory instruments of ANEEL (Agencia Nacional de Energia Eletrica) Regulatory Agency have led to the necessity of a taxonomy that can adapt for this important strategic sector. In this thesis, it is proposed a taxonomy and model of error sequence process for assessment of human error specifically designed to meet the context of operation and maintenance of Hydroelectric Power System. To illustrate the new taxonomy it was collected and analyzed data from about ten years of human error records related to the generation and transmission of Hydroelectric Power Company in Brazil. It was collected 605 reports by human error shutdown from 1998 to 2009. A BBN-Base methodology for the quantification of human error is also discusses. The taxonomy, model for error sequence process as well as the BBN-Based model are illustrated via an example of application in the context of the Brazilian Hydroelectric Power Industry.