In der vorliegenden Arbeit wird eine emirische Untersuchung vorgestellt, in der mittels eines multimethodalen Zuganges eine Anforderungsanalyse für intermediäre Tätigkeiten umgesetzt wurde. Genutzt wurden zur Anforderungsanalyse Experteniterviews, eine Anwendung der Grid-Technik, sowie die Critical-Incidence-Technique.

Dans le domaine de la distribution d’électricité, l’arrivée progressive de production décentralisée rend certains réseaux de distribution exportateurs de puissance, au point où des investissements sont nécessaires pour permettre l’évacuation de la puissance produite. La progression de l’instrumentation des réseaux permet au distributeur d’obtenir des informations de plus en plus riches sur la production décentralisée et la question d’intégrer celles-ci dans les procédés de planification français se pose. À partir de relevés de production, on vérifie que les situations de référence utilisées pour dimensionner le réseau présentent un risque d’occurrence suffisant pour nécessiter des investissements. Si des solutions offrant de la flexibilité sont utilisées, celles-ci peuvent être intégrées au processus de planification. Une étude technico-économique est nécessaire et des méthodes sont proposées pour estimer les fréquences de forte production et la dispersion possible pour un ou plusieurs producteurs. Des modèles simples de loi jointe sont proposés. Le cas pratique de l’effacement de production est testé sur des réseaux considérés réalistes. À partir des relevés réels et des modèles, l’effacement est comparé à l’option d’investir au niveau du poste source. Des abaques de décision sont tracés permettant une projection rapide du distributeur. Par ailleurs dans une optique d’utilisation de l’effacement pour gérer des contraintes intra-réseau, une méthode de calcul de load flow probabilisé est proposée ; celle-ci permet d’estimer rapidement la quantité d’effacement requise et de réaliser un comparatif économique entre plusieurs options. / In the field of electricity distribution, some grids are more and more frequently exporters due to the gradual arrival of decentralized generation. Some grid investments are required to allow the evacuation of the power generated. As more and more information about decentralized production are available for the DSO, the question of integrating them into the planning processes arises.From data of energy production, it appears that the current situations tested for sizing the grid are likely to be reached on a several years basis. If consumption or production flexibilities are used, these can be integrated into the planning process. A technical and economic study is needed and methods are proposed to estimate the frequencies of high production and possible dispersion for one or more producers. Simple models of joint distribution are proposed.The practical case of curtailing production has been tested on networks with realistic scenarios of producers arrival. From actual data and previous models, curtailment and reinforcement are compared on a technico-economic basis. A abacus is plotted as a simple decision tool for the DSO. The question of using curtailment to defer or avoid intra-grid investment has also been investigated ; a calculation method of probabilistic load flow is proposed; it gives a quick and accurate estimation of the energy to curtail in order to avoid a constraint.

Outages and power interruptions are a common and unenviable part of power distribution system operations. Growing demands on reliability in distribution systems has opened up for new technological solutions for fault location at MV and LV level in distribution systems, previously reserved for transmission systems. This report compiles and compares available methods for fault location at distribution level and maps the current fault location process at the power distribution company Ellevio, with the aim of reaching a recommendation for a new fault location scheme. The advocated method is an impedance based method motivated by its reliability, applicability and affordability. The performance and implementation procedure is evaluated through a number of case studies where the methods impact on power reliability demonstrated as well as the need for grid analysis before implementation. Fault indicators and fault current, through relay communications, was identified as key factors for a successful implementation of the method.

L’augmentation de la consommation pour soutenir la croissance, le souci de réduction des gaz à effet de serre et les avancées technologiques ont favorisé le développement des sources d’énergie renouvelables depuis les années 90. L’implantation de ces générateurs décentralisés a progressivement modifié l’architecture du réseau en passant d’un modèle vertical à une situation davantage clusterisée. Ce réseau maillé voit ainsi apparaitre de nouveaux acteurs, à la fois producteurs et consommateurs (en anglais, les « prosumers»). Pour ce type de structures, la stratégie classique actuelle consiste à acheter l'ensemble de l'énergie consommée alors que la totalité de la production est vendue séparément à des tarifs intéressants. Avec les progrès réalisés sur les différentes technologies de stockage, de nouveaux degrés de liberté apparaissent et des opérations plus intelligentes deviennent possibles. L’objet de l’étude est un microréseau comprenant un générateur photovoltaïque et un consommateur tertiaire associés à un moyen de stockage. Deux technologies sont envisagées avec des volants d’inertie dans un premier temps et une batterie électrochimique (Li-ion) par la suite. Les domaines d’étude relatifs à ce type de système sont la gestion énergétique par planification, la commande temps réel et le dimensionnement. Les travaux de cette thèse se concentrent d’abord sur la problématique de gestion par optimisation des flux d’énergie. Différents algorithmes sont ainsi utilisés et comparés pour planifier le fonctionnement du microréseau. L’objectif est de diminuer la facture énergétique en tenant compte des données de consommation et production mais également des politiques tarifaires en vigueur et d’éventuelles contraintes de fonctionnement imposées par le fournisseur d’énergie. Dans un second temps la problématique de dimensionnement est abordée avec une démarche de conception optimale intégrant la boucle de gestion dès la phase de design. Nous montrons plus particulièrement comment l’adéquation entre les méthodes d’optimisation utilisées et le modèle du microréseau employé peut permettre la réduction significative des temps de calcul. Une configuration optimale du microréseau, valable sur des horizons temporels longs intégrant les alternances saisonnières, peut finalement se dégager. Les travaux se concluent sur une phase d’analyse avec des dimensionnements établis pour différents contextes tarifaires. Le but est de dégager des domaines permettant de valoriser et justifier l’installation d’un moyen de stockage qui s’avère indispensable pour soutenir le développement des sources d’énergies renouvelables et assurer la transition énergétique. / To face the increasing demand of electrical power in compliance with the liberalization of the electricity market and the need of reducing CO2 emissions, many distributed energy resources have emerged and especially the generation systems that utilize renewable energy sources. In the nearfuture, the grid could be described as an aggregation of several microgrids both consumer and producer. For those "prosumers", a classical strategy consists in selling all the highly subsidized production at important prices while all consumed energy is purchased. Smarter operations now become possible with developments of energy storage technologies and evolving prices policies. The microgrid considered in the thesis is composed of an industrial load and a photovoltaic generator associated to an energy storage. Two technologies are considered with high speed flywheels on one hand and a Li-ion electrochemical battery on the other. The common study referring to such systems allude to the optimal scheduling, the real-time management and the sizing methodology. Firstly in the thesis, the optimal power flow dispatching is performed using various algorithms. Those operations aim at reducing the electrical bill taking account of consumption and production forecasts as well as the different fares and possible constraints imposed by the power supplier. Then the design strategy is investigated. The approach consists in simultaneously integrating the energy management and the sizing of the system. We particularly underline the complexity of the resulting optimization problem and how it can be solved using suitable optimization methods in compliance with relevant models of the microgrid. We specifically show the reduction of the computational time allowing the microgrid simulation over long time durations in the optimization process in order to take seasonal variations into account. In the last part a cost analysis is performed, and different design are computed depending on the prices policies. The goal is to determine a financial context that would encourage the deployment of storage systems that are necessary to favor the development of intermittent renewable energy sources.

Made available in DSpace on 2014-06-11T19:29:40Z (GMT). No. of bitstreams: 0 Previous issue date: 2011-07-01Bitstream added on 2014-06-13T18:39:29Z : No. of bitstreams: 1 franco_pb_me_sjrp.pdf: 2600940 bytes, checksum: c5d3454406b1b9e853e5a555f0a171cb (MD5) / Diversos são os esforços para o desenvolvimento de políticas de escalonamento em grid computacional. O uso de simuladores de grid computacional é de especial importância para o estudo de algoritmos de escalonamento de tarefas. Através dos simuladores, é possível avaliar e comparar o desempenho de diferentes algoritmos em diferentes cenários. Apesar das ferramentas de simulação fornecerem funcionalidades básicas para simulação de ambientes distribuídos, elas não disponibilizam políticas internas de escalonamento de tarefas; além disso, a implementação dos algoritmos deve ser feita pelo próprio usuário. Portanto, o objetivo deste trabalho é desenvolver a biblioteca de escalonamento de tarefas LIBTS (Library Tasks Scheduling) e adaptá-la ao simulador SimGrid para oferecer aos usuários uma ferramenta que possibilite o estudo de algoritmos de escalonamento de tarefas em grid computacional. Através da LIBTS os usuários podem comparar os algoritmos implementados (Workqueue (WQ), Workqueue with Replication (WQR), Sufferage, XSufferage, Dynamic FPLTF) em diversos cenários, como também desenvolver e implementar novos algoritmos de escalonamento de tarefas. Este trabalho também proporciona uma revisão de literatura sobre grid computacional, apresentando as características e metodologias dos algoritmos implementados na LIBTS e as principais características das ferramentas de simulação. Além disso, os cenários de testes criados para comparar os algoritmos validaram o funcionamento da biblioteca e o funcionamento correto dos algoritmos na LIBTS / Too much has been done to develop scheduling policies in computational grid. The use of computational grid simulators is particularly important for studying the algorithms of task scheduling. Through the simulators it’s possible to assess and compare the performance of different algorithms in various scenarios. Despite the simulation tools provide basic features for simulation in distributed environments, they don’t offer internal policies of task scheduling, so that the implementation of the algorithms must be realized by the user himself. Therefore, this study aims to develop the library of task scheduling LIBTS (Library Tasks Scheduling) and adapt it to the SimGrid simulator to provide the users with a tool to analyze the algorithms of task scheduling in the computational grid. Through the LIBTS, the users can compare the implemented algorithms (Workqueue (WQ), Workqueue with Replication (WQR), Sufferage, XSufferage, Dynamic FPLTF) in several scenarios, as well as to develop and implement new algorithms of task scheduling. This work also provides a literature review about the computational grid, presenting the characteristics and methodologies of the implemented algorithms in the LIBTS and the most important features of the simulation tools. Furthermore, the test scenarios created to compare the algorithms validate the library operation and the correct operation of the algorithms in LIBTS

[pt] O desenvolvimento de metaheurísticas paralelas autonômicas para serem executadas eficientemente em ambientes de grid é o objetivo desta tese. A aplicação paralela deve ser capaz de se auto-adaptar às mudanças que ocorrem dinamicamente no ambiente, sem que o usuário precise interferir diretamente no código da mesma. Para isso, a metaheurística autonômica deve ser vista como uma aplicação com dois níveis independentes: middleware e estratégia. O middleware é responsável por gerenciar todo o ambiente de execução, de acordo com as características da aplicação. A estratégia hierárquica distribuída permite a cooperação entre todos os processos envolvidos, sem degradar o desempenho da aplicação devido ao aumento da comunicação entre processos. Para validar esta proposta foram desenvolvidas duas implementações paralelas de metaheurísticas, uma para o problema do torneio com viagens espelhado e a outra para o problema da árvore geradora de custo mínimo com restrição de diâmetro. Para ambos os problemas, as implementações desenvolvidas foram testadas no ambiente grid Sinergia, formado por máquinas localizadas em três diferentes cidades do Estado do Rio de Janeiro. As parelizações foram capazes de melhorar, para várias instâncias, os melhores resultados conhecidos na literatura. / [en] The development of autonomic parallel metaheuristics to be efficiently executed in computational grid is the challenge of this thesis. The parallel application must be able to self-adjust to the changes that occur dynamically in the environment, without the user needing to interfere directly in the code of the application. For this, the autonomic metaheuristic should be seen as an application on two independent levels: middleware and strategy. The middleware is responsible for managing the entire execution environment, according to the characteristics of the application. The distributed hierarchical strategy enables the cooperation between all processes involved, without degrading the performance of the application due to increased communication between processes. To validate this proposal, two parallel implementations of metaheuristics were developed, one for the mirrored traveling tournament problem and the other for the diameter constrained minimum spanning tree problem. For both problems, the developed implementations were tested in the grid Synergy environment, formed by machines located in three different cities in the state of Rio de Janeiro. The paralelizations improved, for several instances, the best known results in the literature.

Pesquisas em sistemas paralelos e distribuídos de alto desempenho apresentam limitações no que se refere a análise, projeto, implementação e execução automática e transparente de aplicações. Essas limitações motivaram o projeto do MidHPC (do inglês Middleware for High Performance Computing, ou seja, Middleware para Computação de Alto Desempenho), que balanceia transparente e automaticamente cargas de trabalho considerando a capacidade dos recursos computacionais e o comportamento das aplicações envolvendo: processamento, acesso a disco, memória e rede. Para utilizar todo o potencial do MidHPC, aplicações devem ser escritas utilizando o modelo de programação concorrente, tal como o padrão POSIX de threads (pthreads). Aplicações desenvolvidas seguindo esse modelo de programação podem ser executadas em ambientes de Grid sem alteração de código fonte ou recompilação. Durante a execução, tarefas de uma mesma aplicação paralela comunicam-se, transparentemente, por meio de um sistema de memória compartilhada distribuída. O objetivo deste trabalho foi desenvolver alguns dos módulos do projeto MidHPC e integrar demais ferramentas que haviam sido previamente desenvolvidas pelo grupo. Este trabalho permite aplicar, em ambientes reais, todos os conceitos de escalonamento de processos estudados e desenvolvidos durante o projeto MidHPC / Current researches on parallel and distributed systems present limitations regarding the analysis, design, implementation and automatic execution of high performance applications. Those limitations motivated the design of MidHPC (Middleware for High Performance Computing), which transparently and automatically balances workloads considering computing resources capacities and application behavior such as: processing, network, memory and disc accesses. In order to use all of the MidHPC potential, applications must be developed following the concurrent programming model, using the POSIX thread standard (pthreads). Applications developed according to this programming model can be executed in the Grid environment with no source code modification nor recompilation. During the execution, tasks of the same application communicate, transparently, using a distributed shared memory system. The objective of this master thesis was to develop modules of the MidHPC project and integrate tools that were previously developed by the research group. This work allows applying, in real-world environments, all process scheduling concepts studied and developed during the MidHPC project

Determinar quais são os aparelhos eletrônicos de uma residência que possuem maior influência na conta de luz não é tarefa trivial. As Redes de Sensores Sem Fios (RSSF) auxiliam os usuários nessa tarefa, permitindo descobrir se há algum tipo de desperdício no ambiente monitorado e assim, auxiliá-los a fazer as devidas correções. Por isso, é fundamental usar nas smart grids métodos que detectam novidades, também conhecido como anomalias , de forma individual e autônoma, para os usuários quando algo anômalo surge no consumo de energia dos equipamentos eletrônicos. Tais anomalias podem surgir, por exemplo, quando um equipamento consome energia acima do esperado, o que pode indicar um defeito. Nesse contexto, este trabalho propõe um método inteligente, nomeado como Novelty Detection Power Meter (NodePM), para detectar as novidades no consumo de energia dos equipamentos eletrônicos monitorados por uma smart grid. O NodePM detecta as novidades considerando a entropia de cada equipamento monitorado, a qual é calculada com base em um modelo de cadeia de markov que é gerado através de um algoritmo de aprendizado de máquina. Para tanto, o NodePM é integrado a uma plataforma de monitoramento remoto de consumo de energia, que consiste de uma RSSF associada a uma aplicação em nuvem. Para validar o desempenho do NodePM foram feitos experimentos utilizando a análise de variância e testes paramétricos e não-paramétricos. Os resultados de tais experimentos, obtidos mediante a análise estatística, evidenciou a viabilidade do NodePM na plataforma desenvolvida / It is not a simple task to determine which pieces of elevtronic equipment have the greatest influence on the electricity bill. The Wireless Sensor Networks (WSN) assist users in this task, allowing to discover if there is any type of a waste in a monitored environment and thus, help them to take proper actions. Hence, it is of crucial importance to use intelligent methods in the smart grids for a novelty detection and to inform the users in an individual and autonomous way when some anomaly has occurred in the energy consumption of electronic equipment. These anomalies can arise, for instance, when a piece of equipment consumes more energy than expected. In this context, we propose an intelligent method, named the Novelty Detection Power Meter (NodePM), to detect the novelties in the energy consumption of electonic equipment monitored by a smart grid. The NodePM detects the novelties considering the entropy of each device monitored, which is calculed based on a Markov chain model that is generated through a machine learning algorithm. For this end, the NodePM is integrated into a platform for the remote monitoring of energy consumption, which consists of a WSN associated with a cloud application. To validate the performance of the NodePM, experiments were done using analysis of variance and parametric and non-parametric tests. The result of these tests, which were obtained from a statistical analysis, provided evidence of the feasibility of the NodePM in the platform that was developed

Orientador: Marco Aurelio Amaral Henriques / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Eletrica e Computação / Made available in DSpace on 2018-08-12T06:09:14Z (GMT). No. of bitstreams: 1 Pereira_AnaMariadeSeixas_M.pdf: 2667642 bytes, checksum: d79dd53fe62b9832524a344ff3c494ee (MD5) Previous issue date: 2008 / Resumo: O gerenciamento e a utilização de ambientes de computação em grade requerem um grande esforço na obtenção das informações necessárias para a administração e para a identificação e resolução de problemas. Acompanhar o desempenho das aplicações, determinar a origem de problemas, identificar e eliminar gargalos são funções que requerem informações detalhadas sobre a plataforma e sobre as aplicações sendo executadas. Neste trabalho apresentamos uma proposta para implementação de uma ferramenta para monitoramento do sistema JoiN de processamento maciçamente paralelo virtual. Após a análise de algumas das ferramentas hoje existentes para monitoramento de ambientes distribuídos não identificamos entre elas uma solução que satisfaça os requisitos do sistema JoiN e, por esta razão, implementamos uma nova ferramenta que procura integrar as melhores características encontradas nas ferramentas analisadas. As principais dificuldades encontradas na implementação desta ferramenta são relacionadas à obtenção e à publicação de um grande número de informações que permitam a observação e o monitoramento do sistema JoiN e que facilitem seu uso e sua administração, interferindo o mínimo possível com seu desempenho. Os resultados obtidos mostram que a ferramenta implementada no sistema JoiN oferece uma boa relação custo/benefício no monitoramento das principais funções deste sistema sem causar impactos significativos na execução de suas aplicações paralelas. / Abstract: Grid Computing environments management and utilization require a great effort in gathering necessary information for site administration and identification/resolution of problems. Discovering applications performance, determining the origin of problems, identifying and eliminating bottlenecks are functions thar require detailed information about the platform and the applications running. In this work we present a proposal for implementation of a monitoring tool for JoiN system, a massively virtual environment for parallel processing. After the analysis of some of the available monitoring tools for distributed environments, we could not identify among them a solution that meets JoiN requirements and, therefore, we decided to implemented a new tool which seeks to incorporate the best features found on analyzed tools. The main difficulties found in the implementation of this tool are related to the collection and publication of a large number of information that allow JoiN observation and monitoring and that helps its administration and use, interfering as little as possible with its performance. The results show that the tool implemented in JoiN system offers a good cost/benefit relationship in monitoring the main functions of this system without causing significant impacts in the execution of parallel applications. / Mestrado / Engenharia de Computação / Mestre em Engenharia Elétrica

Emerging Software Defined Networking (SDN) technology has provided excellent flexibility to large-scale networks in terms of control, management, security, and maintenance. On the other hand, recent years witnessed a tremendous growth of the critical infrastructure networks, namely the Smart-Grid, in terms of its underlying communication infrastructure. Such large local networks requires significant effort in terms of network management and security. We explore the potential utilization of the SDN technology over the Smart Grid communication architecture. Specifically, we introduce three novel SDN deployment scenarios in local networks of Smart Grid. Moreover, we also investigate the pertinent security aspects with each deployment scenario along with possible solutions. On the other hand, we conducted experiments by using actual Smart Grid communication data to assess the recovery performance of the proposed SDN-based system. The results show that SDN is a viable technology for the Smart Grid communications with almost negligible delays in switching to backup wireless links.