abstract: Many core modern multiprocessor systems-on-chip offers tremendous power and performance optimization opportunities by tuning thousands of potential voltage, frequency and core configurations. Applications running on these architectures are becoming increasingly complex. As the basic building blocks, which make up the application, change during runtime, different configurations may become optimal with respect to power, performance or other metrics. Identifying the optimal configuration at runtime is a daunting task due to a large number of workloads and configurations. Therefore, there is a strong need to evaluate the metrics of interest as a function of the supported configurations. This thesis focuses on two different types of modern multiprocessor systems-on-chip (SoC): Mobile heterogeneous systems and tile based Intel Xeon Phi architecture. For mobile heterogeneous systems, this thesis presents a novel methodology that can accurately instrument different types of applications with specific performance monitoring calls. These calls provide a rich set of performance statistics at a basic block level while the application runs on the target platform. The target architecture used for this work (Odroid XU3) is capable of running at 4940 different frequency and core combinations. With the help of instrumented application vast amount of characterization data is collected that provides details about performance, power and CPU state at every instrumented basic block across 19 different types of applications. The vast amount of data collected has enabled two runtime schemes. The first work provides a methodology to find optimal configurations in heterogeneous architecture using classifiers and demonstrates an average increase of 93%, 81% and 6% in performance per watt compared to the interactive, ondemand and powersave governors, respectively. The second work using same data shows a novel imitation learning framework for dynamically controlling the type, number, and the frequencies of active cores to achieve an average of 109% PPW improvement compared to the default governors. This work also presents how to accurately profile tile based Intel Xeon Phi architecture while training different types of neural networks using open image dataset on deep learning framework. The data collected allows deep exploratory analysis. It also showcases how different hardware parameters affect performance of Xeon Phi. / Dissertation/Thesis / Masters Thesis Engineering 2019

Practicing engineers need an integrated building, HVAC and control simulation tool for optimum HVAC design and retrofit. Various tools are available to the researchers, but these are not appropriate for the consulting engineer. To provide the engineer with a tool which can be used for typical HVAC projects, new models for building, HVAC and control simulation are introduced and integrated in a user-friendly, quick-to-use tool. The new thermal model for buildings is based on a transfer matrix description of the heat transfer through the building shell. It makes provision for the various heat flow paths that make up the overall heat flow through the building structure. The model has been extensively verified with one hundred and three case studies. These case studies were conducted on a variety of buildings, ranging from a 4m2 bathroom, to a 7755 m2 factory building. Eight of the case studies were conducted independently in the Negev Desert in Israel. The thermal model is also used in a program that was custom-made for the AGREMENT Board (certification board for the thermal performance of new low-cost housing projects). Extensions to the standard tool were introduced to predict the potential for condensation on the various surfaces. Standard user patterns were incorporated in the program so that all the buildings are evaluated on the same basis. In the second part of this study the implementation of integrated simulation is discussed. A solution algorithm, based on the Tarjan depth first-search algorithm, was implemented. This ensures that the minimum number of variables are identified. A quasi-Newton solution algorithm is used to solve the resultant simultaneous equations. Various extensions to the HVAC and control models and simulation originally suggested by Rousseau [1] were implemented. Firstly, the steady-state models were extended by using a simplified time-constant approach to emulate the dynamic response of the equipment. Secondly, a C02 model for the building zone was implemented. Thirdly, the partload performance of particular equipment was implemented. Further extensions to the simulation tool were implemented so that energy management strategies could be simulated. A detailed discussion of the implications of the energy management systems was given and the benefits of using these strategies were clearly illustrated, in this study. Finally, the simulation tool was verified by three case studies. The buildings used for the verification ranged from a five-storeyed office and laboratory building, to a domestic dwelling. The energy consumption and the dynamics of the HVAC systems could be predicted sufficiently accurately to warrant the use of the tool for future building retrofit studies / Thesis (PhD)--University of Pretoria, 1997. / gm2014 / Mechanical and Aeronautical Engineering / unrestricted

Tesis para optar al grado de Magíster en Ciencias de la Ingeniería, Mención Eléctrica / Se define una microrred como una agrupación de cargas y recursos energéticos distribuidos que funciona como un único sistema controlable, capaz de operar en paralelo o aislado de la red eléctrica. Las microrredes son proveedores de energía locales que pueden reducir los gastos de energía, reducir las emisiones, aumentar la confiabilidad y son alternativas de energización emergentes. El correcto uso de sus recursos energéticos disponible permite lograr una operación más eficiente en una microrred, por ejemplo; reducir sus costos, mejorar ingresos, alargar la vida útil de los equipos y limitar el impacto ambiental. Algunos de estos objetivos se contraponen y es por esto que es necesario compensarlos para obtener el mejor despacho energético. Por esta razón el uso de un sistema de gestión de energía para microrredes cobra gran importancia. En este trabajo se desarrollaron modelos matemáticos y luego se implementaron en una herramienta computacional para el despacho energético óptimo de microrredes, con énfasis en tres aspectos. Primero, los servicios complementarios que una microrred puede ofrecer: arbitraje de energía, reducción de emisiones, reducción de potencia punta, reserva de potencia en giro y ofertas de reducción de consumo. Segundo, un modelo de almacenamiento de baterías enfocado en seis fenómenos: envejecimiento cíclico y calendario, la ley de Peukert, la pérdida de capacidad, autodescargas y la limitación de carga/descarga. Tercero, se incluye un módulo maestro-esclavo para lidiar con la estocasticidad ante problemas intempestivos en la red, manteniendo así la confiabilidad de la microrred cuando se aísla, aun si esta ofrece servicios. Estos tres aspectos son integrados en un modelo de programación lineal entera mixta para el despacho óptimo de una microrred, minimizando los costos de operación y reinversión. En el presente trabajo, se simulan la operación de tres microrredes reales bajo diferentes escenarios cada uno. El primer caso es la microrred aislada de Huatacondo, el segundo es la microrred conectada de CIGRE y el tercero es la microrred conectada de la cárcel de Santa Rita. Los resultados obtenidos muestran reducción en los costos de hasta 4.3% en la microrred de Huatacondo, hasta 2.9% para CIGRE y hasta 7% para Santa Rita al considerar servicios y utilizando un modelo detallado de almacenamiento. En el caso de la microrred aislada de Huatacondo, la reducción se basó principalmente en la extensión de la vida útil del banco de baterías. Para las dos microrredes conectadas los servicios más atractivos fueron ofrecer sus capacidades flexibles no utilizadas a la red. Esto considera servicios como reducción de consumo, reducción de demanda punta o reserva en giro. Servicios enfocados en transferencia de altos volúmenes de energía, como el arbitraje de energía, no fueron atractivos dado el costo asociado al uso de equipos de almacenamiento.

Blockchain has become the most revolutionary technology in the 21st century. In recent years, one of the concerns of world energy isn't just sustainability yet, in addition, being secure and reliable also. Since information and energy security are the main concern for the present and future services, this thesis is focused on the challenge of how to trade energy securely on the background of using distributed marketplaces that can be applied. The core technology used in this thesis is distributed ledger, specifically blockchain. Since this technology has recently gained much attention because of its functionalities such as transparency, immutability, irreversibility, security, etc, we tried to convey a solution for the implementation of a secure peer-to-peer (P2P) energy trading network over a suitable blockchain platform. Furthermore, blockchain enables traceability of the origin of data which is called data provenience. In this work, we applied a secure blockchain technology in peer-to-peer energy sharing or trading system where the prosumer and consumer can trade their energies through a secure channel or network. Furthermore, the service management functionalities such as security, reliability, flexibility, and scalability are achieved through the implementation. \\ This thesis is focused on the current proposals for p2p energy trading using blockchain and how to select a suitable blockchain technique to implement such a p2p energy trading network. In addition, we provide an implementation of such a secure network under blockchain and proper management functions. The choices of the system models, blockchain technology, and the consensus algorithm are based on literature review, and it carried to an experimental implementation where the feasibility of that system model has been validated through the output results.

京都大学 / 新制・課程博士 / 博士(工学) / 甲第23203号 / 工博第4847号 / 新制||工||1757(附属図書館) / 京都大学大学院工学研究科電気工学専攻 / (主査)教授 引原 隆士, 教授 土居 伸二, 教授 梅野 健 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

Wireless power transfer is a method of transferring electric power from a transmitter to a receiver without requiring any physical connection between the two. Dynamic Wireless Power Transfer (DWPT) entails having the transmitters buried under the roadway and the receiver unit being installed on the Electric Vehicle (EV). In this method, EVs are charged while driving over the transmitters as they receive bursts of electric energy at the time of significant alignment between transmitters and receivers. Compared to the stationary charging method which involves parking the EV for long hours for a full charge, the dynamic charging method (i.e., DWPT) offers convenience as the vehicle gets charged while driving. It also facilitates extended driving range of EVs. Despite offering these advantages, DWPT causes a few significant issues. DWPT charging results in a transient power profile both at grid side and EV side, which not only hampers grid-side regulation but also affects EV-battery longevity. To address these two issues, both grid-side and EV-side energy management are needed to be employed to protect the grid and the vehicle from sudden exposure to harmful power transients. In this dissertation, the grid-side and EV-side energy management methods have been investigated. Firstly, a detection system to safely detect the vehicle on charging lane is proposed. This detection system is used to facilitate safe and efficient operation of DWPT chargers on EV roadways. Secondly, A novel DWPT system is proposed, which reduces the grid-side power transients with minimal additional hardware requirements. Finally, an EV-side energy management system is proposed which reduces the exposure of EV batteries to pulsating DPWT-power, thereby helping batteries to last longer.

The objective of this work is to design and implement an architectural framework for a web-based demand management system that allows an electric utility to reduce system peak load by automatically managing end-use appliances based on homeowners' preferences. The proposed framework comprises the following components: human user interface, home energy management (HEM) algorithms, web services for demand response communications, selected ZigBee and smart energy profile features for appliance interface, and security aspects for a web-based HEM system. The proposed web-based HEM system allows homeowners to be more aware about their electricity consumption by allowing visualization of their real-time and historical electricity consumption data. The HEM system enables customers to monitor and control their household appliances from anywhere with an Internet connection. It offers a user-friendly and attractive display panel for a homeowner to easily set his/her preferences and comfort settings. An algorithm to autonomously control appliance operation is incorporated in the proposed web-based HEM system, which makes it possible for residential customers to participate in demand response programs. In this work, the algorithm is demonstrated to manage power-intensive appliances in a single home, keeping the total household load within a certain limit while satisfying preset comfort settings and user preferences. Furthermore, an extended version of the algorithm is demonstrated to manage power-intensive appliances for multiple homes within a neighborhood. As one of the demand response (DR)-enabling technologies, the web services-based DR communication has been developed to enable households without smart meters or advanced metering infrastructure (AMI) to participate in a DR event via the HEM system. This implies that an electric utility can send a DR signal via a web services-enabled HEM system, and appropriate appliances can be controlled within each home based on homeowner preferences. The interoperability with other systems, such as utility systems, third-party Home Area Network (HAN) systems, etc., is also taken into account in the design of the proposed web services-based HEM system. That is, it is designed to allow interaction with authorized third-party systems by means of web services, which are collectively an interface for machine-to-machine interaction. This work also designs and implements device organization and interface for end-use appliances utilizing ZigBee Device Profile and Smart Energy Profile. Development of the Home Area Network (HAN) of appliances and the HAN Coordinator has been performed using a ZigBee network. Analyses of security risks for a web-based HEM system and their mitigation strategies have been discussed as well. / Master of Science

Indiana University-Purdue University Indianapolis (IUPUI) / This thesis proposes an Adaptive Rule-Based Energy Management Strategy (ARBS EMS) for a parallel hybrid electric vehicle (P-HEV). The strategy can effciently be deployed online without the need for complete knowledge of the entire duty cycle in order to optimize fuel consumption. ARBS improves upon the established Preliminary Rule-Based Strategy (PRBS) which has been adopted in commercial vehicles. When compared to PRBS, the aim of ARBS is to maintain the battery State of Charge (SOC) which ensures the availability of the battery over extended distances. The proposed strategy prevents the engine from operating in highly ineffcient regions and reduces the total equivalent fuel consumption of the vehicle. Using an HEV model developed in Simulink, both the proposed ARBS and the established PRBS strategies are compared across eight short duty cycles and one long duty cycle with urban and highway characteristics. Compared to PRBS, the results show that, on average, a 1.19% improvement in the miles per gallon equivalent (MPGe) is obtained with ARBS when the battery initial SOC is 63% for short duty cycles. However, as opposed to PRBS, ARBS has the advantage of not requiring any prior knowledge of the engine effciency maps in order to achieve optimal performance. This characteristics can help in the systematic aftermarket hybridization of heavy duty vehicles.

Energy Management strategy for sustainable regional development has been selected as the topic of my research due to the fact that energy demand alongside with energy dependency have been continuously growing from a long term perspective. Sustainable development is defined by three imperatives – energy efficiency, ecology and security. Review of the current state and analysis of historical trends in Energetics at global and regional level are covered in this research. Results of the Multi-Criteria Decision Analysis introduce a set of implications and recommendations for Energy Management strategy in the Czech Republic.

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第18530号 / 情博第534号 / 新制||情||95(附属図書館) / 31416 / 京都大学大学院情報学研究科知能情報学専攻 / (主査)教授 岡部 寿男, 教授 松山 隆司, 教授 阿久津 達也 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM