Master of Science / Department of Architectural Engineering and Construction Science / Fred L. Hasler / Typical building power supplied from power plants has significantly evolved over the last century. After power is generated and transferred from the power plant, it is distributed to the customer. The concept of Microgrid was introduced to address increasing concerns with power reliability requirements for some facilities. The Microgrid is a localized electric grid that can operate autonomously from the traditional electric grid (Macrogrid). Compared with generator sets, a Microgrid provides a faster system response and recovery to either whole or part of the electric load for a facility. The Microgrid can operate under two operation conditions: grid-tied mode and island mode. When it is working individually like an “island”, the system is not providing or receiving disturbance to or from the Macrogrid. The purpose of this paper is to give a detailed introduction of Microgrid and present research and conclusion about its feasibility. This report references previously published research to explain what a Microgrid is. Also, two detailed case studies provide a discussion about the feasibility of the Microgrid in terms of its reliability, economics and environmental impact - air quality. Although there are many challenges that Microgrids are facing, there are quite a number of reasons to consider them. The goal is to balance the benefits and challenges of Microgrids depend on each case. No doubt, the existing power grid will still provide the majority of power supply for global population. However, many companies and government-funded laboratories are investing time and money into research and development of Microgrids. With the advancement of the Microgrids, it is likely that Microgrids will be playing a larger role in providing secure, reliable energy to the building industry.

Η παρούσα διπλωματική εργασία έχει ως αντικείμενο την μελέτη και την παρουσίαση των στρατηγικών για τον έλεγχο της ροής ισχύος στα μικροδίκτυα. Επιπροσθέτως γίνεται μια εκτενής παρουσίαση της δομής και των αρχών λειτουργίας των μικροδικτύων καθώς και των δυνατοτήτων που έχουμε για κατανεμημένη παραγωγή και αποθήκευση και των ηλεκτρονικών μετατροπέων ισχύος που χρησιμοποιούμε για τον έλεγχο αυτών / This thesis regards to the subject of studying and presentation of power control strategies regarding power flow in Microgrid systems. Additionally is a detailed presentation of the structure and operation principles of microgrids and the possibilities we have for distributed generation and storage and power electronic converters used to control them.

No description available.

The need for a continuous supply of electric power is vital to providing the basic services of modern life. The energy infrastructure that the vast majority of the world depends on, while very reliable, is also very vulnerable. This infrastructure is particularly vulnerable to disruptions caused by natural disasters. Interruptions of electric service can bring an end to virtually all the basic services that people are dependent on. Recent natural disasters have highlighted the vulnerabilities of large, economically developed, regions to disruptions to their supply of electricity. The widespread devastation from the 2011 Japanese Tsunami and Hurricane Irene in North America, have demonstrated both the vulnerability of the contemporary power grids to long term interruption of service and also the potential of microgrids to ride through these interruptions. Microgrids can be used before, during, and after a major natural disaster to supply electricity, after the main grid source has been interrupted. This thesis researches the potential of clean energy microgrids for disaster recovery. Also a model of a proposed residential microgrid for transient analysis is developed. As the world demands more energy at increasingly higher levels of reliability, the role of microgrids is expected to grow aggressively to meet these new requirements. This thesis will look at one potential application for a microgrid in a residential community for the purpose of operating in an independent island mode operation. / Master of Science

L'évolution du stockage d'énergie permet de développer des méthodes innovantes de gestion de l'énergie à une échelle locale. Les micro réseaux électriques sont une forme émergente de petits réseaux électriques munis de production locale, de stockage d'énergie et en particulier d'un système de gestion de l'énergie (EMS pour Energy Management System). De nombreuses études et recherches scientifiques ont été menées pour proposer diverses stratégies d'implémentation de ces EMS. Néanmoins il n'existe pas à ce jour d'articulation claire et formelle de ces méthodes permettant leur comparaison. L'une des principales difficultés pour les EMS, est la gestion des dynamiques des différents systèmes énergétiques. Les variations de courant vont à la vitesse de l'électron, la production d'énergie solaire photovoltaïque varie au gré des nuages et différentes technologies de stockages peuvent réagir plus ou moins vites à ces phénomènes imprévisibles. Nous étudions dans ce manuscrit, un formalisme mathématique et des algorithmes basés sur la théorie de l'optimisation stochastique multi-étapes et la Programmation Dynamique. Ce formalisme permet de modéliser et de résoudre des problèmes de décisions inter-temporelles en présence d'incertitudes, à l'aide de méthodes de décomposition temporelle que nous appliquons à des problèmes de gestion de l'énergie. Dans la première partie de cette thèse, "Contributions à la décomposition temporelle en optimisation stochastique multi-étapes", nous présentons le formalisme général que nous utilisons pour décomposer en temps les problèmes d'optimisation stochastique avec un grand nombre de pas de temps. Nous classifions ensuite différentes méthodes de contrôle optimal au sein de ce formalisme. Dans la seconde partie, "Optimisation stochastique de stockage d'énergie pour la gestion des micro réseaux", nous comparons différentes méthodes, introduites dans la première partie, sur des cas réels. Dans un premier temps, nous contrôlons une batterie ainsi que des ventilations dans une station de métro récupérant de l'énergie de freinage des trains, en comparant quatre algorithmes différents. Dans un second temps, nous montrons comment ces algorithmes pourraient être implémentés sur un système réel à l'aide d'une architecture de contrôle hiérarchique de micro réseaux électrique en courant continu. Le micro réseaux étudié connecte cette fois ci de l'énergie photovoltaïque à une batterie, une super-capacité et à une charge électrique. Enfin nous appliquons le formalisme de décomposition par blocs temporels présenté dans la première partie pour traiter un problème de gestion de charge de batterie mais aussi de son vieillissement long terme. Ce dernier chapitre introduit 2 algorithmes basés sur la décomposition par blocs temporels qui pourraient être utilisés pour le contrôle hiérarchique de micro réseaux ou les problèmes d'optimisation stochastique présentant un grand nombre de pas de temps. Dans la troisième et dernière partie, "Logiciels et expériences", nous présentons DynOpt.jl un paquet développé en langage Julia qui a permis de développer toutes les applications de cette thèse et bien d'autres. Nous étudions enfin l'utilisation de ce paquet dans un cas de pilotage réel de système énergétique : la gestion intelligente de la température dans une maison de l'équipement Sense City / The development of energy storage paves the way to innovative methods to manage energy at a local scale. Micro grids are a novel kind of electrical grids with local production (renewable and waste energy), local demand, local storage and an Energy Management System (EMS). A wide literature already studies EMS implementations in micro grids but the produced methods are not exhaustively framed and compared. One of the main difficulty in micro grids energy management is to handle the different dynamics of electrical devices. Current variations are lighting fast, solar power changes quickly, different kind of storage react at different paces and batteries ageing is a slow process. We studya mathematical framework and algorithms, based on multistage stochastic optimization theory and Dynamic Programming, to model and solve energy management problems in micro grids with time decomposition methods. In the first part of this thesis, Contributions to time decomposition in multistage stochastic optimization, we present a general framework to decompose temporally large scale stochastic optimization problems into smaller subproblems. We then classify multiple existing resolution methods inside this framework. In the second part, Stochastic optimization of energy storage for management of micro grids, we compare different methods presented in the first part on realistic applications. First we control a battery and a ventilation in a subway station recovering subways braking energy with four different algorithms. Then we present how these results could be implemented on a real micro grid. We implement a fast online control method to stabilize the voltage in a simulated islanded DC micro grid connecting solar panels, an electrical load and two sorts of energy storage: a battery and a supercapacitor. Finally we apply our time decomposition framework to a problem of long term aging and energy management of a storage in a micro grid. This last chapter introduces a framework to model time decomposition of micro grids hierarchical control architectures, as well as two algorithms to solve temporally large scale stochastic optimization problems.In the third part, Softwares and experimentations, we present DynOpt.jl, a Julia language package developed to produce all the results of this thesis and more. Then we study an application of this software to the control of a real test bed: the energy aware temperature regulation of a real house in the equipment named "Sense City"

A blackout is defined as the loss of electric power for a given period in a particular area. With increasing dependence on reliable electric power, the social and economic ramifications of blackouts are dire, negatively impacting the productivity, safety, and security of communities. To reduce blackout occurrence, power system planners incorporate redundancy and advanced controls to the grid to make it more adaptable to disturbances. However, adding redundant transmission lines is not only expensive, it is suboptimal in some contexts. While it is unattainable to have no blackout, it is possible and necessary to implement measures that minimize the likelihood and scale of these outages. This work proposes a solution that uses a microgrid with advanced energy storage and solar PV to mitigate blackouts in Kigali, the capital of Rwanda. A description and steady state analysis of major weaknesses in the Rwandan electric grid is presented. A microgrid application capable of islanding from the system is simulated in the steady state and shown to strengthen the system and decrease the likelihood of blackouts in Kigali. The composition of the microgrid is then designed, simulated, and optimized for technical and financial feasibility using the HOMER model. A microgrid that uses energy storage and solar PV is shown to not only be feasible, but also competitive with current costs of electricity in Rwanda. For comparison, different combinations that include diesel generation are also simulated. / Master of Science

The electric power grid forms the foundation for several other critical infrastructures of national importance such as public health, transportation and telecommunication systems, to thrive. The current power grid runs on the century-old technology and faces serious challenges of the 21st century - Ever-increasing demand and the need to provide a sustainable way to meet the growing demand, increased requirement of resilience against man-made and natural disasters, ability to defend against cyber attacks, increasing demand for reliable power, requirement to integrate with alternate energy generation and storage technologies. Several countries, including the United States, have realized the immediate need to modernize the grid and to pursue the goal of a smart grid. Majority of recent grid modernization efforts are directed towards the distribution systems to be able to meet these new challenges. One of the key enablers of a fully functional Smart Grid are microgrids — subsystems of the grid, utilizing small generation capacities at the distribution system level to increase the overall reliability and power quality of the local grid. It is one of the key directions recommended by national electric delivery technologies roadmap in United States as well as policy makers for electricity delivery in many countries. Microgrids have witnessed serious research activity in the past few years, especially in areas such as multi-agent system (MAS) architectures for microgrid control and auction algorithms for microgrid electricity transaction. However, most of the prior research on electricity transaction in microgrids fails to recognize and represent the true nature of the microgrid electricity market. In this research, a comprehensive microgrid electricity market has been designed, taking into account several unique characteristics of this new market place. This thesis establishes an economic rationale to the vision of wide-scale deployment of microgrids serving residential communities in near future and develops a comprehensive understanding of microgrid electricity market. A novel concept of Community Microgrids is introduced and the market and business models for electricity transaction are proposed and validated based on economic forecasts of key drivers of distributed generation. The most important contribution of this research deals with establishing a need for a trustworthy model framework for microgrid market and introducing the concept of reputation score to market participants. A framework of day-ahead energy market (DAEM) for electricity transaction, incorporating an approach of using the reputation score to incentivize the sellers in the market to be trustworthy, has been designed and implemented in MATLAB with a graphical user interface (GUI). Current implementation demonstrates a market place with two sellers and nine buyers and is easily scalable to support multiple market participants. The proposed microgrid electricity market may spur the deployment of residential microgrids, incorporating distributed generation, thereby making significant contribution to increase the overall reliability and power quality of the local grid. / Master of Science

Les convertisseurs multicellulaires DC-DC sont utilisés dans de nombreuses applications et de nombreux systèmes électriques. Ils présentent un intérêt particulier pour des applications spécifiques liées aux énergies renouvelables et aux Microgrids. Leur principal avantage provient de leur capacité intrinsèque à réduire les ondulations liées au découpage des grandeurs électriques en entrée et en sortie du système de conversion. Cette propriété intéressante au niveau système peut être étendue au fonctionnement interne du convertisseur en adjoignant à ce dernier un élément de filtrage par inductances couplées magnétiquement. Ce composant permet d’étendre les propriétés externes de réduction des ondulations au fonctionnement de chaque cellule du convertisseur. Il permet également d’augmenter la dynamique propre du système de conversion. Ces propriétés permettent de réduire significativement le niveau et le volume de filtrage en entrée et sortie du convertisseur et donc d’augmenter de manière importante sa compacité et son rendement énergétique. Cependant, l’ajout de ce dispositif magnétique induit, de par le couplage des équations du système qu’il provoque, une complexification du contrôle de la structure associée également à la nécessité d’augmenter le nombre de capteurs.Ce travail de thèse a pour objectif d’établir et d’évaluer différents modes de contrôle pour les convertisseurs multicellulaires DC-DC. Le point commun aux méthodes proposées est de permettre la gestion aussi bien des grandeurs externes au convertisseur que des grandeurs internes constituées par les courants de circulation entre cellules connectées en parallèle. Ces composantes de courant sont également nommées « courants différentiels ». Trois types de contrôle sont étudiés : Pour le premier, des correcteurs linéaires classiques sont utilisés conjointement avec des techniques de découplage des équations du système. La robustesse de ces méthodes de contrôle vis-à-vis des incertitudes sur la connaissance des paramètres du système fait l’objet d’un focus particulier dans cette partie du travail. Pour le second, une version modifiée de la technique de commande connue sous le nom Model Predictive Control est proposée. Celle-ci permet d’assurer le contrôle de la fréquence de commutation et l’entrelacement des commandes PWM des cellules. Pour le troisième mode, nous étudions une méthode basée sur le contrôle vectoriel direct des courants différentiels.Une implantation sur un système numérique équipé d’un micro-processeur et d’un FPGA est proposée et permet de valider les résultats de l’étude théorique. / The interleaved multicell DC-DC power converters are broadly used in many applications and systems especially in renewable energy systems and microgrids. They reduce the current ripple at the input and output side. Also, an implemented magnetic coupling between cells leads to reduce the current ripple in each of them and to improve the dynamical electrical behavior. These properties involve a reduction on the filtering requirements and so, allow to improve the converter compactness as well as its conversion efficiency. Nevertheless, for such power converters, the control complexity is also increased as well as the number of required sensors.The thesis aims to establish different mode of control of interleaved multicell DC-DC converters. The common point of these methods is to control the external quantities at the output of the converter but also the internal quantities, constituted by the circulating currents between parallel cells or in other words the differential currents. Three main strategies are investigated: the first one uses classical linear controllers with different decoupling technics and focuses on the robustness regarding the system parameters variations. The second one uses a Model Predictive Control technic which is designed to provide a fix switching frequency and interleaving of the cells PWM commands. The last one presents a space vector direct control of the differential currents.In a last part, these control principles are tested on a prototype and implemented on a Microcontroller and FPGA board in order to carry out an experimental verification.

Tesis para optar al grado de Doctor en Ingeniería Eléctrica / The integration of large numbers of Distributed Energy Resources (DERs) into the distribution system could take place either by reinforcement of the existing network assets, or the incorporation of active management of flexible resources into different sections of the distribution network. For active management of a distribution network, the design of control strategies is necessary for an efficient and reliable large-scale integration of DERs. Besides the benefit of supporting the use of renewable energy sources, DERs play an important role in improving the resilience and sustainability of the electricity distribution system and also in the generation of new market opportunities. In this thesis, the active management of DERs is proposed using a hierarchical energy management system (EMS) applied to "Energy Communities". Energy communities are a concept which allows different end users to cooperate in their energy interactions with the aim of maximising their self-consumption, minimising energy costs, reducing peak power levels or a combination of these and other beneficial goals as well. The hierarchical EMS proposed allows incorporating mechanisms to ensure both the realisation of short-term power balancing objectives and long-term energy management, benefiting the microgrid owner and the distribution network operator. The hierarchical EMS is designed in two levels: main grid level and microgrid level. At the microgrid level, a real-time local rule-based controller is proposed and at the higher level, a Robust model predictive control (MPC) is used to manage the uncertainty associated with renewable distributed generation and electricity demand. The uncertainty is incorporated into the Robust MPC controller based on fuzzy prediction interval models in order to help the system to be prepared for errors in the predictions that might yield sub-optimal decisions. Several case studies are used to test the performance of the hierarchical EMS for the operation and coordination of microgrids. Robust EMS based on fuzzy prediction interval models is compared to the deterministic EMS and with a basic EMS without energy storage system (ESS). The results show that the deterministic and Robust EMSs provide improvements over the case without ESS, as they offer mechanisms for efficient energy management. The incorporation of an ESS into the energy community benefits both the end user, by reducing energy cost, and the distribution network operator, by limiting the peak power levels and enabling increased penetration of distributed generation (DG). Additionally, the hierarchical EMS is able to keep the community power flow close to the reference power defined by the higher level controller with minimum energy cost, among other benefits. Finally, end users operating as Energy Communities can optimise the use of DG and the size of the ESS required.

Cargas e fontes elétricas de pequeno porte, integradas através de alimentadores de distribuição de baixa tensão, agrupadas e gerenciadas de tal maneira que se comportem como uma única entidade controlável dentro de um sistema de potência e com a capacidade de operar de forma paralela ou isolada da rede elétrica convencional, constituem-se em um novo tipo de sistema elétrico conhecido como Minirrede. Esta tese aborda o problema relacionado à estabilidade dinâmica de Minirredes com controle autônomo de frequência e tensão, quando operadas de forma isolada da rede convencional. As minirredes analisadas são compostas por combinações de fontes elétricas convencionais e fontes alternativas que não geram naturalmente em 50 ou 60 Hz, sendo necessária a utilização de inversores para sua conexão com a rede elétrica. A análise se realiza no espaço de estados com metodologias baseadas nas teorias que envolvem os autovalores e autovetores da matriz de estado. Para isto, modelos no espaço de estados são desenvolvidos para cada componente da Minirrede, mantendo um compromisso entre precisão e complexidade. O caráter analítico da análise permite investigar a relação entre a estabilidade da Minirrede e os parâmetros do sistema, tais como ganhos dos controladores, dimensionamento da rede, configurações de geração, entre outros. Análises complementares de sensibilidade dos autovalores aos parâmetros elétricos do sistema permitem inferir o comportamento dinâmico de diversas configurações de Minirredes, obtendo-se importantes conclusões à respeito. Os resultados analíticos são verificados através de simulação computacional no ambiente Simulink/Matlab®. / Electrical loads and small power sources, integrated through low voltage distribution feeders, managed in such a way to behave as one controllable entity within in a power system, and with the ability to operate connected or non-connected to the conventional power system is a new kind of electrical system known as Microgrid. This thesis addresses the problem related to dynamic stability of island Microgrids with autonomous control of frequency and voltage. The Microgrids analyzed are composed of combinations of conventional and alternative power sources that do not generate electricity naturally in 50 or 60 Hz, therefore inverters are needed to provide AC network interface. The analysis is performed in the state space with methodologies based on theories involving the eigenvalues and eigenvectors of the state matrix. For this, state-space models are developed for each component of Microgrid, maintaining a compromise between accuracy and complexity. The analytical nature of this study allows investigating the relationship between the stability of Microgrid and parameters of the system, such as controller gains, network design, generation architectures, among others. Complementary sensitivity analyzes of eigenvalues to the electrical parameters of the system allow us to infer the dynamic behavior of various configurations of Microgrids, obtaining important conclusions on the matter. The analytical results are verified by computer simulation using the platform Simulink/Matlab®.