Die vorliegende Dissertation untersucht die Stabilität und Regelung von Stromnetzen mit hohem Anteil Erneuerbarer Energien (EE). Dabei stehen drei Forschungsfragen, zu den neuartigen Herausforderungen für die zukünftige Stromnetzstabilität im Zuge der Energiewende, im Vordergrund. Erstens soll untersucht werden wie die Resilienz von Stromnetzen gemessen und im zweiten Schritt auch verbessert werden kann. Dabei zeige ich den notwendigen Detailgrad für transiente Stabilitätsuntersuchungen auf. Die zweite Frage lautet wie, trotz des zunehmenden Ausbaus von EE in Verteilnetzen, die statische Spannungsstabilität garantiert und Leitungsüberlastungen verhindert werden können. Hierfür analysiere ich mit einem konzeptionellen hierarchischen Verteilnetzmodell das zukünftige Potential für die Erzeugung von Blindleistung aus dezentralen Ressourcen am Beispiel Deutschlands. Die dritte Frage, wie eine dynamisch-stabile Integration von EE möglich ist, bildet den Hauptfokus meiner Dissertation. Dabei untersuche ich wie neuartige dynamische Aspekte EE, wie intermittente Fluktuationen oder auch Mess- und Reaktionszeiten von Leistungselektronik, die dynamische Netzstabilität beeinflussen und wie mögliche Instabilitäten durch Konzepte der Nachfragesteuerung behoben werden können. Dabei stoße ich bei der Analyse lokaler intermittenter Fluktuationen in ohmschen Verteilnetzen auf ein bemerkenswertes Wechselspiel zwischen Eigenschaften der Netzdynamik und -topologie. Als Zweites zeige ich wie mit der Einführung von Leistungselektronik und den damit verbundenen Mess- und Reaktionszeiten Resonanzkatastrophen hervorrufen werden können. Schließlich präsentiere ich wie die dezentrale Nachfragesteuerung von Elektroautos dynamische Instabilitäten, hervorgerufen durch Fluktuationen von EE, bereinigen kann. Zusammenfassend behandelt diese Arbeit verschiedene Aspekt zur Stabilität zukünftiger Stromnetze und integriert dabei sukzessive neuartige dynamische Aspekte von EE. / This PhD thesis is centered around the "Stability and Control of Power Grids with high Renewable Energy Share". With a conceptual modelers approach, I tackle three overarching questions related to the novel challenges the energy transition poses for the stability of future power grids. The first question focuses on how to measure and subsequently improve the resilience of a power grid. Here, I contribute important insights on the necessary model detail for transient stability assessments. The second question concerns how to ensure static voltage stability and avoid capacity overloading while the deployment of Renewable Energy Sources (RES) in the distribution grid layers is massively increasing. As a possible solution to this problem I analyze the future technical potential of reactive power provision from decentral resources in Germany. The third question, and main focus of this thesis, is on how to integrate renewable energies in a dynamically stable way. Specifically, I investigate the influence of intermittent RES and measurement delays from power electronic resources on frequency stability and how the latter can be restored by concepts of demand control. First, for local intermittent fluctuations in lossy distribution grids I find a remarkable and subtle but robust interplay of dynamical and topological properties, which is largely absent for lossless grids. Second, I show how delays may induce resonance catastrophes and how the existence of critical delays sets an upper limit for measurement times. Third and last, I present how the right parameterization of decentral electric vehicle control can completely overcome issues of short-term dynamic instability related to RES fluctuations. This control avoids demand synchronization and high battery stress. Altogether, this thesis investigates the stability of future power grids moving towards integrating more aspects of renewable energy dynamics. Finally, I point out open questions to encourage further research.

Diese Arbeit ist der Frage gewidmet, wie die Stabilität eines Stromnetzes mit seiner Topologie zusammenhängt. Stabilität wird hier mit Hilfe von Bassin-Stabilität vermessen, einer nichtlokalen und nichtlinearen Methode, mit der sich quantifizieren lässt, wie stabil ein Stromnetz gegen große Störungen ist. Netzartige Systeme werden mit den Methoden der Theorie komplexer Netzwerke beschrieben. Bassin-Stabilität wird zunächst formal eingeführt. Dann wird die Methode angewendet, und zwar einmal auf abstrakte und einmal auf konkrete Weise. Der abstrakte Ansatz ist dadurch motiviert, dass die Funktion eines Stromnetzes auf dem synchronen Zusammenwirken seiner Komponenten beruht. Er betrifft die Frage, wie generelle dynamische Netzwerke strukturiert sein sollten, um stabile Synchronisation zu unterstützen. Es stellt sich heraus, dass Netzwerke mit maximaler Bassin-Stabilität am entgegengesetzten Ende eines gewissen netzwerktheoretischen Spektrums angesiedelt sind als Netzwerke, die optimal sind in Hinsicht auf lokale Stabilität. Dies deutet darauf hin, dass in der Entwicklung von synchron funktionierenden Systemen die Optimierung auf lokale Stabilität hin und die gleichzeitige Optimierung auf nichtlokale Stabilität hin als entgegengesetzte Kräfte gewirkt haben. Der konkrete Ansatz geht aus von einem Stromnetzmodell aus der Ingenieuersliteratur. Bassin-Stabilität wird komponentenweise eingesetzt, um zu untersuchen, wie der Grad der Stabilität eines Stromnetzes gegen große Einzelknoten-Störungen beeinflusst wird von gewissen Mustern in der Netztopologie. Simulationen in einem Ensemble von Stromnetzen bringen diverse Statistiken hervor, die alle eine wesentliche Beobachtung stützen: Ausgerechnet das kostengünstigste und weitverbreitetste Anschlussschema - bestehend aus Stichleitungen - vermindert die Netzstabilität beträchtlich. Es lässt sich ein fundamentales Designprinzip ableiten: Vermeide Stichleitungen. / This thesis addresses the question how the stability of a power grid depends on the network topology. Stability is assessed here by means of basin stability, a nonlocal and nonlinear inspection scheme that allows to quantify how stable a power grid is against large perturbations. Networked systems are characterized using the instruments of complex network theory. A rigorous definition of basin stability is provided. Then, basin stability is applied once in an abstract and once in a more applied way. The abstract approach is motivated by the fact that power grids rely on the synchronous operation of their components, and focusses on the problem how general dynamical networks should be structured to support a stable synchronous state. It turns out that networks with maximum basin stability are located at the opposite end of a certain network-theoretical spectrum than networks optimal with respect to local stability. This suggests that, during the evolution of synchronizing networks, the optimization for local stability and the simultaneous optimization for nonlocal stability have acted as opposing forces. The more applied approach draws on a power grid model from the engineering literature. A component-wise version of basin stability is used to assess how a grid''s degree of stability against large single-node perturbations is influenced by certain patterns in the wiring topology. Simulations in an ensemble of power grids yield various statistics that all support one main finding: The widespread and cheapest of all connection schemes, so-called dead ends and dead trees, strongly diminish stability. The inverse is also true: `Healing'' dead ends by addition of transmission lines substantially enhances stability. This indicates a basic design principle for future power grids: avoid dead ends.

Energy consumption in the form of fuel or electricity is ubiquitous globally. Among energy types, electricity is crucial to human life in terms of cooking, warming and cooling of shelters, powering of electronic devices as well as commercial and industrial operations. Therefore, effective prediction of future electricity consumption cannot be underestimated. Notably, repeated imbalance is noticed between the demand and supply of electricity, and this is affected by different weather profiles such as temperature, wind speed, dew point, humidity and pressure of the electricity consumption locations. Effective planning is therefore needed to aid electricity distribution among consumers. Such effective planning is activated by the need to predict future electricity consumption within a short period and the effect of weather variables on the predictions. Although state-of-the-art techniques have been used for such predictions, they still require improvement for the purpose of reducing significant predictive errors when used for short-term load forecasting. This research develops and deploys a near-zero cooperative probabilistic scenario analysis and decision tree (PSA-DT) model to address the lacuna of significant predictive error faced by the state-of-the-art models and to analyse the effect of each weather profile on the cooperative model. The PSA-DT is a machine learning model based on a probabilistic technique in view of the uncertain nature of electricity consumption, complemented by a DT to reinforce the collaboration of the two techniques. Based on detailed experimental analytics on residential, commercial and industrial data loads, the PSA-DT model with weather profiles outperforms the state-of-the-art models in terms of accuracy to a minimal error rate. This implies that its deployment for electricity demand planning will be of great benefit to various smart-grid operators and homes. / School of Computing / M. Sc. (Computer Science)

The objective of the proposed research is to analyze automated residential energy management technology using primary energy source utilization. A residential energy management system (REMS) is an amalgamation of hardware and software that performs residential energy usage monitoring, planning, and control. Primary energy source utilization quantifies power system levels impacts on power generation cost, fuel utilization, and environmental air pollution; based on power system generating constraints and electric load. Automated residential energy management technology performance is quantified through a physically-based REMS simulation. This simulation includes individual appliance operation and accounts for consumer behavior by stochastically varying appliance usage and repeating multiple simulation iterations for each simulated scenario. The effect of the automated REMS under varying levels of control will be considered. Aggregate REMS power system impacts are quantified using primary energy source utilization. This analysis uses a probabilistic economic dispatch algorithm. The economic dispatch algorithm quantifies: fuel usage and subsequent environmental air pollution (EAP) generated; based on power system generating constraints and electric load (no transmission constraints are considered). The analysis will comprehensively explore multiple residential energy management options to achieve demand response. The physically-based REMS simulation will consider the following control options: programmable thermostat, direct load control, smart appliance scheduling, and smart appliance scheduling with a stationary battery. The ability to compare multiple automated residential energy management technology options on an equal basis will guide utility technology investment strategies.

The world faces the unprecedented challenge of the need change to a new energy era. The introduction of distributed renewable energy and storage together with transportation electrification and deployment of electric and hybrid vehicles, allows traditional consumers to not only consume, but also to produce, or store energy. The active participation of these so called "prosumers", and their interactions may have a significant impact on the operations of the emerging smart grid. However, how these capabilities should be integrated with the overall system operation is unclear. Intelligent energy management systems give users the insight they need to make informed decisions about energy consumption. Properly implemented, intelligent energy management systems can help cut energy use, spending, and emissions. This thesis aims to develop a consumer point of view, user-friendly, intelligent energy management system that enables vehicle drivers to plan their trips, manage their battery pack and under specific circumstances, inject electricity from their plug-in vehicles to power the grid, contributing to frequency regulation.

Submitted by Johnny Rodrigues (johnnyrodrigues@ufcg.edu.br) on 2018-05-01T14:10:40Z No. of bitstreams: 1 NATÃ VENÂNCIO DE MELO - DISSERTAÇÃO PPGCC 2015..pdf: 1756988 bytes, checksum: 57d7a292fbbca336c632f27a98fe34bc (MD5) / Made available in DSpace on 2018-05-01T14:10:40Z (GMT). No. of bitstreams: 1 NATÃ VENÂNCIO DE MELO - DISSERTAÇÃO PPGCC 2015..pdf: 1756988 bytes, checksum: 57d7a292fbbca336c632f27a98fe34bc (MD5) Previous issue date: 2015-08-19 / Análise de causa raiz de ocorrências de falha é uma tarefa extremamente importante para a recomposição e estabilidade do Sistema Elétrico de Potência (SEP). Nos Centros de supervisão e Controle, os operadores têm o desafio de analisar centenas de alarmes e descobrir a causa raiz das falhas num período curto de tempo. Caso não sejam executadas ações emergenciais, pode ocorrer um efeito cascata na rede elétrica e, em última instância, prejuízos de diversas ordens. surge, portanto, a necessidade de soluções que automatizem a descoberta da causa raiz de falhas. este trabalho apresenta uma nova técnica de análise de causa raiz que oferece a geração de árvores de propagação de falha, adaptabilidade e mudanças topológicas, facilidade de adaptação a novos alarmes, tratamento à dessincronização de alarmes e meios fáceis de alertar as regras de negócio. ele se destaca pela simplicidade de implementação, que utiliza regras de causa e efeito entre os equipamentos, e pelo bom desempenho na geração de árvores de falha. Sua validação foi realizada por especialistas da área de operação de sistemas elétricos de potência, utilizando ocorrências de falhas reais decorrentes de grandes redes elétricas. / The root cause analysis of fault is an extremely task to keep the stability of the Eletric Power System (EPS). In the Supervision and Control Centers, the operators have the challenge of analyzing hundreds of alarms and discover the root cause of failures in a short period of time. If actions are not performed very quickly, the fault can spread and generate a ripple effect in the power grid. To help the operators, it becomes necessary to develop a solution to automate the discovery of root cause of faults. This work presents a new technique of root cause analysis of fault that offers the generation of trees of fault propagation, adaptability to topological changes, facility to change the business rules. It shows a simple implementation that use rules of cause and effect between the equipments and a good performance in generating fault trees. the new technique was validated by engineers of the area of eletric poert systems with fault occurrences of big power grids.

As redes inteligentes ou smart grids surgiram como um dos principais agentes dentro da reestruturação que o sistema elétrico de potência vem sofrendo há alguns anos. Mais especificamente, essa evolução vem tomando destaque nas redes de distribuição de energia elétrica. O surgimento de novos equipamentos, com tecnologia moderna, permite que o sistema seja monitorado em tempo real e manobrado de forma remota mediante as necessidades. Com o objetivo de otimizar cada vez mais as funcionalidades desses equipamentos, aplicando e promovendo sistemas capazes de funcionar de forma autônoma é que estudos vem desenvolvendo novas metodologias para atender as necessidades de recomposição das redes de distribuição diante à alguma anomalia. Um novo conceito que surgiu como vertente das redes inteligentes vem movendo o setor acadêmico e as concessionárias responsáveis pela distribuição de energia chamado reconfiguração de redes de distribuição, vem sendo tema de discussões e estudos sobre as melhores alternativas de implantação. Essa dissertação apresenta um modelo para utilização em sistemas baseado em reconfiguração, que identifique os melhores resultados para o conjunto quando é necessário fazer a transferência de carga de uma subestação de energia por perda em suas fontes de alimentação. O modelo leva em conta os principais índices de qualidade e continuidade de energia obtidos a partir de um método de busca exaustiva selecionando as melhores opções de manobras para reconfigurar os alimentadores mediante a uma anomalia no sistema de distribuição. O método multicritério PROMETHEE é utilizado para definir as melhores sequências de manobra na rede e consequentemente os melhores desempenhos para o sistema após o remanejamento das cargas. O modelo apresenta uma característica um pouco diferente dos que se encontra na literatura, pois não utiliza nenhum método de compensação para melhorar os índices de níveis de tensão e perdas de energia. A metodologia proposta foi avaliada e validade mediante a simulações em uma rede real pertencente a uma concessionária de energia elétrica e seus resultados apresentaram uma melhora significativa para o sistema de distribuição. / Smart grids have emerged as one of the key players in restructuring the electric power system has been suffering for some years. More specifically, this evolution has taken prominence in the electricity distribution networks. The emergence of new equipment with modern technology allows the system to be monitored in real time and operated remotely by the needs. In order to optimize more and more features of the equipment, applying and promoting systems that function autonomously is that research is developing new methodologies to meet the needs of restoration of distribution networks on the abnormality. A new concept that has emerged as component of smart grids has been moving the academic sector and the utilities responsible for distribution of energy called reconfiguration of distribution networks, has discussions and studies theme being on the best deployment of alternatives. This thesis presents a model for use in reconfiguring based systems, identifying the best results for the set when you need to load transfer of a loss of power substation in their power supplies. The model takes into account the main quality indicators and energy continuity obtained from an exhaustive search method selecting the best options maneuvers to reconfigure feeders by an anomaly in the distribution system. The multicriteria method PROMETHEE is used to define the best maneuver sequences on the network and therefore the best performance for the system after the relocation of cargo. The model presents a characteristic slightly different from that found in the literature, it does not use any compensation method for improving the rates of voltage levels and energy losses. The proposed methodology was evaluated and validity by simulations in a real network belonging to an electric utility and the results showed a significant improvement to the distribution system.

Este trabalho apresenta uma análise da utilização da comunicação de banda estreita via rede elétrica (Power Line Communication - PLC) em ambiente industrial. Para análise, utilizou-se um kit de desenvolvimento para aplicações baseadas em PLC de acordo com o padrão G3-PLC. Ensaios experimentais e resultados são apresentados para diferentes cenários que simulam diversas situações existentes em um ambiente industrial. Demonstra-se a influência da inserção de ruídos no canal PLC, através do acionamento de inversores e motores, afim de verificar a influência desses elementos no desempenho da comunicação PLC. Como base comparativa, utilizou-se a distância máxima de transmissão normalizada da rede ASInterface, muito difundida em ambientes industriais. Recomendações relativas à técnica de modulação e tamanho do pacote de dados são apresentadas. Como resultado final, conclui-se que o padrão de comunicação G3-PLC e sua técnica de modulação ROBO atendem os requisitos de comunicação industrial, principalmente para aplicações de monitoramento de variáveis industriais, onde o tempo não é crítico. / This work presents an analysis of the narrowband communication through the power line (PLC) in an industrial environment. For analysis, a development kit for PLCbased applications according to the G3-PLC standard was used. Experimental tests and results for different scenarios that simulate various situations existing in an industrial environment. An influence of the insertion of noise in the PLC channel is demonstrated by the drive of inverters and motors, in order to verify a nonperformance influence of the PLC communication. As a comparative basis, a maximum standard transmission distance of the AS-Interface network was used, very widespread in industrial environments. Recommendations regarding the modulation technique and the size of the data packet are presented. As a final result, it is concluded that the G3-PLC communication standard and its ROBO modulation technique meet the requirements of industrial communication, especially for industrial variable monitoring applications where time is not critical.

Este trabalho apresenta uma análise da utilização da comunicação de banda estreita via rede elétrica (Power Line Communication - PLC) em ambiente industrial. Para análise, utilizou-se um kit de desenvolvimento para aplicações baseadas em PLC de acordo com o padrão G3-PLC. Ensaios experimentais e resultados são apresentados para diferentes cenários que simulam diversas situações existentes em um ambiente industrial. Demonstra-se a influência da inserção de ruídos no canal PLC, através do acionamento de inversores e motores, afim de verificar a influência desses elementos no desempenho da comunicação PLC. Como base comparativa, utilizou-se a distância máxima de transmissão normalizada da rede ASInterface, muito difundida em ambientes industriais. Recomendações relativas à técnica de modulação e tamanho do pacote de dados são apresentadas. Como resultado final, conclui-se que o padrão de comunicação G3-PLC e sua técnica de modulação ROBO atendem os requisitos de comunicação industrial, principalmente para aplicações de monitoramento de variáveis industriais, onde o tempo não é crítico. / This work presents an analysis of the narrowband communication through the power line (PLC) in an industrial environment. For analysis, a development kit for PLCbased applications according to the G3-PLC standard was used. Experimental tests and results for different scenarios that simulate various situations existing in an industrial environment. An influence of the insertion of noise in the PLC channel is demonstrated by the drive of inverters and motors, in order to verify a nonperformance influence of the PLC communication. As a comparative basis, a maximum standard transmission distance of the AS-Interface network was used, very widespread in industrial environments. Recommendations regarding the modulation technique and the size of the data packet are presented. As a final result, it is concluded that the G3-PLC communication standard and its ROBO modulation technique meet the requirements of industrial communication, especially for industrial variable monitoring applications where time is not critical.

Orientador: Luiz Carlos Pereira da Silva / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Elétrica e de Computação / Made available in DSpace on 2018-08-25T18:20:32Z (GMT). No. of bitstreams: 1 Alcantara_MarcioVenicioPilar_D.pdf: 32763443 bytes, checksum: de62ea1f9e48329f99dd1eb8b0ae1724 (MD5) Previous issue date: 2014 / Resumo: A aplicação das tecnologias de Rede Inteligente (RI) no Setor de Energia Elétrica (SEE) vem sendo bastante discutida e estudada nos últimos dez anos, em especial no Brasil nos últimos cinco anos, havendo diversas iniciativas tanto por parte das empresas de energia elétrica, quanto por parte do governo e da academia, mas ainda permanece uma lacuna quanto ao entendimento da melhor forma de se integrar o consumidor final nessas perspectivas. Dessa forma, do ponto de vista de gerenciamento do consumo/geração de energia seria possível racionalizar de forma automatizada o uso da energia elétrica, e nesse contexto surgiria a possibilidade de se atender parte da demanda através da Conservação de Energia (CE). Esta tese apresenta um estudo do estado da arte acerca de RI no Brasil e no Mundo e sua aplicação na CE, elencando elementos sobre as novas tecnologias e as diversas formas de aplicação das mesmas para promoção da CE. O texto aponta a necessidade de que as iniciativas existentes no Brasil em termos de Leis, Regulações e Planejamento no âmbito da Eficiência Energética no SEE passe a levar em conta a aplicação de novas tecnologias de controle da carga e de microgeração e minigeração no seu planejamento, em especial que leve em consideração a possibilidade de penetração ampla de tecnologias de geração distribuída. É feita a avaliação da potencialidade da ferramenta de simulação GridLAB-D para simulação dos conceitos associados a RI e CE. Realiza-se uma análise do impacto na mudança do modelo regulatório atual de distribuição de energia elétrica, baseado em remuneração pelo aumento do consumo, para um modelo baseado em remuneração pelo aumento da CE, apontando caminhos para uma nova regulação / Abstract: Application of Smart Grid (SG) technologies in the Power Sector has been extensively studied and discussed in the last ten years, especially in Brazil in the last five years, with various initiatives both by electric utilities, as from government and academia. However, remains a gap in the understanding of how best to integrate the end consumer in these perspectives. Thus, from the standpoint of power management/power generation would be possible to automated rationalize the use of electricity, and in this context arise the possibility to meet part of the demand through the Energy Conservation (EC). This thesis presents a study on the state of the art of SG in Brazil and in the World and its application in EC, listing information about new technologies and different ways of applying them to promote the EC. The text points out the need for existing initiatives in Brazil in terms of Laws, Regulations and Planning under the Energy Efficiency in the Power Sector pass to take into account the application of new technologies for load control in its planning in particular taking into account the possibility of large penetration of distributed generation technologies. It is evaluated the capability of GridLAB-D simulation tool for simulation of the concepts associated with SG and EC. Carried out an impact analysis of the change in the current regulatory model for the distribution of electricity, based on compensation by increased consumption, to a model based on compensation by increased EC, indicating paths for new regulation / Doutorado / Energia Eletrica / Doutor em Engenharia Elétrica