Distributed infrastructures (e.g., water networks and electric Grids) are difficult to manage due to their scale, lack of accessibility, complexity, ageing and uncertainties in knowledge of their structure. In addition they are subject to loads that can be highly variable and unpredictable and to accidental events such as component failure, leakage and malicious tampering. To support in-field operations and central management of these infrastructures, the availability of consistent and up-to-date knowledge about the current state of the network and how it would respond to planned interventions is argued to be highly desirable. However, at present, large-scale infrastructures are “data rich but knowledge poor”. Data, algorithms and tools for network analysis are improving but there is a need to integrate them to support more directly engineering operations. Current ICT solutions are mainly based on specialized, monolithic and heavyweight software packages that restrict the dissemination of dynamic information and its appropriate and timely presentation particularly to field engineers who operate in a resource constrained and less reliable environments. This thesis proposes a solution to these problems by recognizing that current monolithic ICT solutions for infrastructure management seek to meet the requirements of different human roles and operating environments (defined in this work as field and central sides). It proposes an architectural approach to providing dynamic, predictive, user-centric, device and platform independent access to consistent and up-to-date knowledge. This architecture integrates the components required to implement the functionalities of data gathering, data storage, simulation modelling, and information visualization and analysis. These components are tightly coupled in current implementations of software for analysing the behaviour of networks. The architectural approach, by contrast, requires they be kept as separate as possible and interact only when required using common and standard protocols. The thesis particularly concentrates on engineering practices in clean water distribution networks but the methods are applicable to other structural networks, for example, the electricity Grid. A prototype implementation is provided that establishes a dynamic hydraulic simulation model and enables the model to be queried via remote access in a device and platform independent manner.This thesis provides an extensive evaluation comparing the architecture driven approach with current approaches, to substantiate the above claims. This evaluation is conducted by the use of benchmarks that are currently published and accepted in the water engineering community. To facilitate this evaluation, a working prototype of the whole architecture has been developed and is made available under an open source licence.

Aviation industry is in many ways considered a pioneer, especially in innovation and implementation of new materials and technologies. The aim of this thesis is to provide comprehensive and complex information on the matter, pointing out the differences in distribution systems and scenarios outlining the expected evolution of these systems in the future. The dissertation defines the basic concepts and the nature of the problem and also talks about the historical development of the GDS to better understand the current ownership and evaluate the importance and status of GDS in the Czech and world market. It analyzes selected GDS and graphically illustrates the positions of the GDS on the certain markets within the category. The basic method used is the comparison which was done by putting the specifics of individual GDS and their subsequent comparison, the thesis also includes SWOT analysis.

Orientador: Walmir de Freitas Filho / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-27T12:29:12Z (GMT). No. of bitstreams: 1 Pinto_AdrianoCosta_M.pdf: 2073375 bytes, checksum: e1c68a8598816ca4909e74ba53dee76d (MD5) Previous issue date: 2015 / Resumo: A crescente demanda por energia elétrica, por vezes em ritmo mais acelerado que os investimentos em expansão das redes de distribuição, tem levado as distribuidoras a operarem próximo aos limites aceitáveis, o que torna toda a operação da rede mais complexa. Um dos desafios atuais é estabelecer um efetivo controle de tensão e potência reativa (Volt/var) na rede buscando melhorar o nível de operação e de eficiência energética da rede. Muitas propostas para encontrar a solução do problema partiram de uma abordagem de forma desacoplada: o controle de tensão e o controle de potência reativa foram resolvidos separadamente. Neste trabalho, porém, foram estudados métodos de solução do problema visando à segurança da operação e à otimização global dos recursos da rede de modo integrado, ou seja, considerando a dependência entre tensão e potência reativa. Na literatura, grande parte dos trabalhos reportam soluções baseadas em modelos elétricos da rede de distribuição. Os métodos estudados nessa dissertação são baseados em técnicas de aprendizado de máquina com o objetivo de construir um modelo capaz de utilizar apenas as medições de tensão e corrente provenientes dos medidores instalados ao longo da rede e obter o melhor despacho dos ajustes dos dispositivos de controle, sem a necessidade de um modelo elétrico do sistema. A grande vantagem de não depender dos dados e modelo elétrico do sistema está associada às imprecisões tipicamente existentes na base de dados elétricos das concessionárias de distribuição de energia elétrica. Neste contexto, primeiramente, propõe-se o uso de aprendizado por reforço, no qual o agente interage com a rede enquanto acumula experiência de operação dos controles. A implementação através do algoritmo Q-Learning permite a construção de um operador virtual da rede de distribuição a partir dos dados provenientes dos medidores instalados em determinadas barras do sistema, dos quais é extraído o estado corrente da condição de carregamento da rede. Os principais aspectos da aplicação do método ao problema de controle integrado de tensão e potência reativa são simulados em redes típicas e as capacidades de aplicação prática ao cenário atual do sistema elétrico são discutidas. Em uma segunda etapa, propõe-se utilizar um algoritmo de aprendizado supervisionado através de Máquinas de Vetores de Suporte (em inglês, Support Vector Machine ¿ SVM), uma técnica eficientemente aplicada a problemas de mineração de dados. O modelo é implementado através de técnicas de classificação, que extraem características relevantes nos conjuntos de dados, a fim de otimizar a operação da rede para cada condição de carregamento, eliminando a necessidade de repetir o treinamento do modelo ou calcular uma nova solução do problema de otimização a cada novo cenário. Discute-se o desempenho do método baseado em SVM para diferentes características de entrada. Investiga-se ainda a generalização do modelo proposto na presença de ruídos nos dados e no caso de reconfiguração da rede. Estudos em sistemas típicos de distribuição mostram que o método proposto é eficiente na solução de problemas práticos do dia-a-dia das concessionárias, principalmente em ambientes com grande volume de dados / Abstract: The growing demand for electricity, sometimes at a faster rate than investments in distribution network expansion, has led utilities operating close to acceptable limits, which makes the network operation more complex. One of current challenges is to establish an effective voltage and reactive power control, improving the operation as well as the efficiency of the distribution network. There are many methods reported to find a solutions for the voltage and reactive power problem. Most of them have adopted a decoupled form, solving the voltage control and reactive power (Volt/var) control separately. However, in this work, methods for the problem solution aiming the operation safety and the global assets optimization are approached in an integrated fashion, i. e., considering the dependence between voltage and reactive power. Most papers reports solution based on electrical models of distribution network. In this dissertation, the methods studied are based on machine learning techniques aiming to build a model with directly power meter data using capability, and set optimal dispatch of controls devices adjustments, without the need of an electrical model of the system and, therefore, not susceptible to inaccuracies of the model of the distribution network under study. Firstly, it proposes a reinforcement learning use, in which the agent interacts with the network while earns control operating experience. The implementation, thought de Q-Learning algorithm allows a construction of a distribution network virtual operation from data obtained from the meters installed on buses. From the meter data, is extracted the current state of the network loading condition. The main aspects of the application of the method to the integrated voltage and reactive power control are simulated in a typical network and the possibilities of practical application in the current scenario of the electrical system are discussed. In a second step, an algorithm for supervised learning via the Support Vector Machine (SVM), a technique applied efficiently to problems in data mining is proposed. The model is implemented by classification techniques, extracting relevant features in the data sets from the power meters in order to optimize the operation of the network for each loading condition. Thus it eliminates the need to retraining model or calculating a new optimization problem solution for each new scenario. Discusses the performance based on different features for SVM model input. Also the generalization capabilities of the proposed model in the presence of noise and in the case of network reconfiguration are studied. Studies in typical distribution systems show that proposed method is a good candidate to solve the practical problem of the system, especially in large networks with large amounts of data / Mestrado / Energia Eletrica / Mestre em Engenharia Elétrica

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-01-19T10:31:58Z No. of bitstreams: 1 gustavojosesantiagorosseti.pdf: 1281297 bytes, checksum: 0d547bba728df4c23bbf1317541f6e39 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-01-25T17:44:12Z (GMT) No. of bitstreams: 1 gustavojosesantiagorosseti.pdf: 1281297 bytes, checksum: 0d547bba728df4c23bbf1317541f6e39 (MD5) / Made available in DSpace on 2016-01-25T17:44:12Z (GMT). No. of bitstreams: 1 gustavojosesantiagorosseti.pdf: 1281297 bytes, checksum: 0d547bba728df4c23bbf1317541f6e39 (MD5) Previous issue date: 2015-08-31 / Esta tese apresenta uma metodologia para maximizar o restabelecimento de cargas em sistema de distribuição de energia elétrica após à ocorrência de uma ou simultâneas contingências. Para tanto, um algoritmo heurístico construtivo é proposto, determinando passo a passo os procedimentos operativos a serem adotados. Aspectos associados com as restrições de radialidade e de tensão nas barras, minimização de manobras de chaves, consumidores prioritários e mínimo corte discreto de carga são considerados a fim de uma representação mais realista do problema. A metodologia é aplicada em sistemas tradicionais da literatura, incluindo um sistema real de médio porte. / This thesis presents a methodology for maximizing the load restoration in power distribution system after simultaneous occurrence of contingency. Therefore, a heuristic constructive algorithm is proposed to determine step by step the operation procedures to be adopted. Aspects associated with the radiality and bus voltage constraints, minimization of maneuvering switches, priority consumers and minimum discrete load shedding are considered to provide a more realistic representation of the problem. The proposed approach is applied in traditional systems from literature including a real medium size test system.

This diploma thesis describes study of interconnection of Letohrad and Lanškroun water distribution systems. The thesis contains an overview of legislative regulations and technical standards related to the construction, design and directional solution of water supply systems. Following that, there is the description of the current state of the affected water mains. Next part is the design of interconnection including hydraulic analysis in using Epanet 2.0 software. The study contains several variants of the solution. The economic assessment is included.

Theoretical part of the diploma thesis purveys information on drinking water and quality requirements, changes of drinking water quality during distribution and accumulation caused by disinfection, chemical processes, corrosion of constructional materials, incrusting solids, biofilms, nitrification of ammonia nitrogen, sediments in pipelines, elution of harmful matter. Attention was given to drinking water quality indicators which are associated with changes in drinking water distribution systems. In experimental part are studied changes of drinking water quality during distribution on the basis of selected drinking water quality indicators. There is specifically chemical oxygen demand, concentration of iron, manganese, ammonia ions, nitrites, nitrates, chloroform and chlorine. For periodical monitoring were selected suitable locality within the framework of Brno distribution system; for illustration were taken water samples from distribution systems of drinking water treatment plants Švařec, Vír, Štítary and Mostiště. For selected methods of analytical determination of above mentioned drinking water quality indicators are presented their characteristics. Obtained information on drinking water quality is mentioned in tabular and graphical form.

A water distribution system (WDS) is based in a network of interconnected hydraulic components to transport the water directly to the customers. Water must be treated in a Water Treatment Plant (WTP) to provide safe drinking water to consumers, free from pathogenic and other undesirable organisms. The disinfection is an important aspect in achieving safe drinking water and preventing the spread of waterborne diseases. Chlorine is the most commonly used disinfectant in conventional water treatment processes because of its low cost, its capacity to deactivate bacteria, and because it ensures residual concentrations in WDS to prevent microbiological contamination. Chlorine residual concentration is affected by a phenomenon known as chlorine decay, which means that chlorine reacts with other components along the system and its concentration decrease. Chlorine is measured at the output of the WTP and also in several considered points within the WDS to control the water quality in the system. Simulation and modeling methods help to predict in an effective way the chlorine concentration in the WDS. The purpose of the thesis is to assess chlorine concentration in some strategic points within the WDS by using the historical measured data of some water quality parameters that influence chlorine decay. Recent investigations of the water quality have shown the need of the use of non-linear modeling for chlorine decay prediction. Chlorine decay in a pipeline is a complex phenomenon so it requires techniques that can provide reliable and efficient representation of the complexity of this behavior. Statistical models based on Artificial Neural Networks (ANN) have been found appropriated for the investigation and solution of problems related with non-linearity in the chlorine decay prediction offering advantages over more conventional modeling techniques. In this sense, this thesis uses a specific neural network application to solve the problem of forecasting the residual chlorine

[pt] Em sistemas de distribuição de energia elétrica, um dos maiores desafios para as distribuidoras é a estimação das perdas técnicas. De acordo com a bibliografia, as perdas elétricas nas redes de distribuição em diferentes países podem variar aproximadamente de 3 porcento e 25 porcento da energia fornecida à rede, o que pode significar grandes impactos nos custos do sistema. Especificamente no Brasil, a adequada avaliação das perdas elétricas fornece informação importante para que o regulador estabeleça as tarifas de distribuição de energia elétrica. Na literatura há diversos métodos para a estimação das perdas técnicas de energia, mas devido à dificuldade na modelagem dos equipamentos do sistema, assim como a falta de informação da energia consumida pelas cargas, as estimações podem acarretar em grandes erros. Para tratar este problema, esta dissertação propõe um novo método baseado em um modelo de carga polinomial modificado para estimar as perdas elétricas, considerando medições de tensão e potência na subestação e, quando disponíveis, medições de tensão e potência demandadas pelas cargas. A contribuição principal do método proposto é o uso da informação da topologia da rede e a correlação entre a potência consumida pelas cargas e as grandezas medidas na subestação. Para detalhar e analisar o desempenho do método proposto são utilizados três sistemas elétricos. Os resultados das estimações são comparados com os resultados obtidos por outros métodos de referência encontradas na literatura e em aplicações práticas. / [en] In electrical distribution systems, one of the greatest challenges for utilities is the estimation of technical losses. According to the literature, energy losses throughout the world s electric distribution networks may vary from country to country approximately between 3 percent and 25 percent of the electricity provided, which may cause great impacts on the electrical system costs. Specifically in Brazil, the appropriate evaluation of the energy losses provides valuable information for the regulator to establish the energy distribution tariffs. In literature, there are different ways for estimating energy losses, but due to the difficulty for modeling precisely the equipment of the system, as well as the lack of information regarding the energy consumed of each load, the energy losses estimation may lead to huge errors. To deal with this problem, it is proposed a new method based on a modified load model, taking into account the measurements of voltages and power at the substation and, when available, the measurements of voltages and power demanded by loads with meters installed. The main contribution of the proposed method is the use of the network information and the correlation between the power consumed by the loads and the voltage and power supplied by the substation. In order to detail and analyze the performance of the proposed method, three electric systems are used. The results of the estimations given by the proposed method are compared to those obtained with other methods found in literature and in practical applications.

[pt] Os recursos disponíveis neste trabalho são a operação do On Load Tap Changer (OLTC) da subestação, possibilidade de cortes de carga e, finalmente, o dimensionamento e despacho de baterias no sistema. Para uma análise mais realista, é abordada, ainda, uma formulação robusta da incerteza da carga e uma representação dos perfis de consumo através de cenários típicos, estabelecidos por agrupamento de similaridade, utilizando algoritmo de mineração de dados K-Means. O sistema teste modificado IEEE 123 barras é empregado na avaliação da metodologia descrita, e indica a viabilidade operacional e econômica da inserção de dispositivos armazenadores de energia no contexto de proposta do trabalho. / [en] The development of studies related to the power applications and economic feasibility of energy storage resources in electricity distribution networks has become promising considering the reduction in the cost of energy storage. Such technology can minimize the intermittence of renewable sources, provide the displacement of peak loads, extend the expansion of the electricity grid infrastructure, among other benefits. In this sense, this dissertation intends to explore and evaluate an integer-mixed linear optimization model, which is originally non-linear, for the planning and operation of energy storage systems inserted in a distribution system that may present unbalanced loads. The model seeks to minimize operation and investment costs, meeting systemic constraints, coordinating the different resources of a distribution system. The resources available in this work are the operation of the On Load Tap Changer (OLTC) of the substation, the possibility of load cuts and, finally, the sizing and dispatch of batteries in the system. For a more realistic analysis, a robust formulation of the load uncertainty and a representation of consumption profiles through typical scenarios, established by similarity clustering, using K-Means data mining algorithm are also addressed. The modified test system IEEE 123 bus is used in the evaluation of the described methodology and indicates the operational and economic feasibility of inserting energy storage devices in the context of the proposed work.

[pt] O cálculo do fluxo de potência em uma rede elétrica consiste em determinar o estado da rede, os fluxos e perdas elétricas nas linhas e as perdas internas totais no sistema. Nesse tipo de problema, a modelagem do sistema é estática e a rede é representada por um conjunto de equações e inequações algébricas. Diferentes métodos de solução foram propostos na literatura para realizar cálculos de fluxo de potência. No entanto, para redes de distribuição, esses métodos devem ser capazes de modelar, com detalhes suficientes, algumas características únicas desses sistemas, como sua estrutura quase radial, a natureza desequilibrada das cargas e a inserção de geradores distribuídos. Além disso, a modelagem do padrão de consumo nos sistemas de distribuição é mais complexa e os parâmetros das linhas são mais difíceis de serem obtidos, quando comparados com o sistema de transmissão. Portanto, a aplicação de métodos tradicionais para cálculos de fluxo de potência em redes de distribuição pode levar a soluções divergentes. Nesse contexto, o presente trabalho propõe uma nova abordagem para cálculos de fluxo de potência em sistemas de distribuição, baseada em Machine Learning. Os modelos propostos utilizam Redes Neurais Artificiais (RNAs) para prever as perdas ativas internas de uma rede de distribuição e os fluxos de potência nas fronteiras com o sistema de transmissão. Simulações numéricas demonstram o desempenho eficiente da abordagem proposta, além de suas vantagens computacionais em relação aos softwares normalmente utilizados nesse tipo de estudo pois, uma vez treinadas, as RNAs podem aproximar, de modo extremamente rápido, cálculos de fluxo de potência, já que apenas operações matriciais são realizadas. Além disso, o trabalho apresenta uma aplicação da metodologia proposta: as previsões, obtidas pela RNA, para os fluxos nas fronteiras com a rede de transmissão foram utilizadas para gerar contratos ótimos de demanda para um sistema de distribuição real no Brasil. / [en] The power flow calculation on an electric network consists of determining the network s state, power flows and electrical losses on the lines, and total losses on the feeder. In this type of problem, the system s modeling is static, and the network is represented by a set of algebraic equations and inequations. Different solution methods were proposed in the literature to perform power flow calculations. However, for distribution networks, these methods must be able to model, with sufficient details, some unique features of these systems, such as their near radial structure, the unbalanced nature of the loads, and distributed generators insertion. Besides that, modeling the consumption pattern in distribution systems is more complex, and the line parameters are more difficult to be obtained when compared to the transmission system. Hence, applying traditional methods for power flow calculations in distribution networks may lead to divergent solutions. Within this context, this work proposes a new approach for power flow calculations in distribution systems based on Machine Learning. The proposed models use Artificial Neural Networks (ANNs) to predict the active internal losses of a distribution network and the power flows at the borders with the transmission system. Numerical simulations demonstrate the effective performance of the proposed approach, as well as its computational advantages over benchmark software programs since, once trained, ANNs can approximate power flow calculations extremely fast, as only matrix operations are needed. Moreover, the work presents an application of the ANN methodology proposed: predictions of the flows at the borders with the transmission network were used to generate optimal demand contracts for a real distribution system in Brazil.