Transformers and storage systems in the electrical grid must be provisioned or sized just as routers and buffers must be sized in the Internet. We prove the formal equivalence between these two systems and use this insight to apply teletraffic theory to sizing the electrical grid, obtaining the capacity region corresponding to a given transformer and storage size. We conduct a fine-grained measurement study of household electrical load. These measurements are essential for two reasons. First, we use them to construct reference models for home loads; these models are used to find the capacity region using the teletraffic theory. Second, these measurements are used in numerical simulations that are done to validate our analysis. More specifically, we compare results of numerical simulations with the results from teletraffic theory. We show not only that teletraffic theory agrees well with numerical simulations but also that it closely matches with the heuristics used in current practice. Moreover, our analysis permits us to develop sizing rules for battery storage electrical grid, advancing the state of the art.

Recently, there have been many initiatives to incorporate advanced controls, two way communications, digital technologies and advanced power system components in the operation and control of power distribution systems. These initiatives are aiming to realize what becomes known as the “Smart grid”. It is expected that a Smart Grid will lead to enhancement in the reliability and efficiency of the power system. The movement towards the Smart Grid is motivated by many factors; the need to integrate more renewable power to mitigate the global warming, the increasing interest in connecting more distributed generation (DG) as a way to postpone large investment in transmission and bulk generation, and the need to increase the reliability of the power system overall to minimize disturbance costs. It is the overall goal of this research to introduce novel distribution system operation techniques to assist in the effort of realizing the “Smart Distribution System” in both normal and system restoration modes. In particular, three main operation functions are dealt with in this research work; Voltage Control, Reactive Power Control and Distribution System Restoration. First for Voltage Control, a reliable and efficient method is proposed to control voltage regulators in order to enable the regulation of multiple feeders with diversified loads using only one regulator provided that no DG is connected to the feeders. Regulator’s tap is selected based on the solution of an integer linear optimization problem. The method has a closed form solution for the optimal tap; that is valuable for real time operation. In addition, necessary condition for feasible solutions is examined. Next, a novel coordinated voltage control scheme is proposed to enable the voltage regulator to efficiently regulate the voltage of multiple feeders in the presence of DGs. The proposed technique is based on placing a Remote Terminal Unit (RTUs) at each DG and each line capacitor. These RTUs coordinate together, through communication, and form a multi-agent system. An important contribution of this research is that the proposed scheme provides the minimum hardware requirement to efficiently estimate the voltage profile of a feeder with DGs. The proposed scheme enables the integration of more DGs into the system by, efficiently, coordinating the operation of voltage regulators and DGs to mitigate voltage rise problem caused by the connection of DGs to the system. Second, for Reactive Power Control, a decentralized reactive power control scheme is proposed to optimally control switched shunt capacitors of the system in order to minimize system losses and maintain acceptable voltage profile. The proposed algorithm provides capacitors with “Advanced Voltage Sensing” capability to enable capacitors to switch in and out according to the global minimum and maximum voltage of the feeder. The proposed technique utilizes the same RTU used for voltage control and relies on the voltage profile estimation technique proposed in this research for the coordinated voltage control. In addition, novel decentralized algorithm is proposed to estimate the feeder voltage profile change as a result of injecting reactive power at the capacitor bus. The proposed reactive power control scheme can be used to coordinate the operation of any number of capacitors connected to the distribution system. Combining voltage control and reactive power control schemes, generalized coordinated voltage control is proposed to coordinate between DGs, shunt capacitors and voltage regulators in order to achieve optimal voltage control for the distribution system and solve the steady state voltage rise problem caused by the connection of DGs, hence, allowing more DGs to be connected to the system. Over and above, the proposed generalized coordinated voltage control enables the realization of a new operation-time DG connection impact assessment concept. Based on this concept, the system will carry out a real-time assessment and decide, based on the available control actions, the maximum DG power that can be allowed to connect to the system at particular operating conditions. This new concept will allow great flexibility to the connection of DGs, most notably, when, due to a change in system configuration, the DG is needed to be connected to a feeder other than the one it was planned for during the planning stage. The last operation function dealt with in this research work is the distribution system restoration. Novel decentralized distribution system restoration scheme is proposed. The proposed scheme is based on dividing the distribution system into zones based on the availability of disconnecting switches. Each zone is controlled by an Agent. The restoration is done based on the coordination between these Agents. Proposed communication protocols between Agents are discussed in details. The goal of the proposed restoration scheme is to maximize the restored power while preserving the radial structure of the distribution system and without exceeding the thermal limit of any equipment in the system. As the proposed technique does not assume any supervision from any central point, this technique will enable the realization of a self-healing distribution system restoration.

Service Orientation (SO) and grid computing are two computing paradigms that when put together using Internet technologies promise to provide a scalable yet flexible computing platform for a diverse set of distributed computing applications. This practice gives rise to the notion of a computing cloud that addresses some previous limitations of interoperability, resource sharing and utilization within distributed computing. <p>In such a Service Oriented Computing Cloud (SOCC), applications are formed by composing a set of services together. In addition, hierarchical service layers are also possible where general purpose services at lower layers are composed to deliver more domain specific services at the higher layer. In general an SOCC is a horizontally scalable computing platform that offers its resources as services in a standardized fashion. <p>Workflow based applications are a suitable target for SOCC where workflow tasks are executed via service calls within the cloud. One or more workflows can be deployed over an SOCC and their execution requires scheduling of services to workflow tasks as the task become ready following their interdependencies. <p>In this thesis heuristics based scheduling policies are evaluated for scheduling workflows over a collection of services offered by the SOCC. Various execution scenarios and workflow characteristics are considered to understand the implication of the heuristic based workflow scheduling.

In order to understand tidal circulation and oceanic current for the seas around Taiwan, this study use a baroclinic unstructured grid model to build a high resolution model. This model use semi-implicit method to solve the dynamic of ocean movement and larger time step can be used to calculate. Unstructured grid can be used to resolve complex coastline and variation of depth. TaiDBMv6 depth data were chosen to describe the depth distribution and grid mesh size were determined by local depth, minimum mesh size is about 0.75 minutes, and maximum 13 minutes. Tidal boundaries use 8 constituents derived from FES2004 and calibrated with 34 tide station records. Data of 2009 were used to evaluate the model results. The average of all station root mean square error was 10.1 cm. Station at east side of Taiwan have smaller errors, which almost lower than 5 cm. The maximum error can be found inside Taiwan Strait, about 25cm, mainly caused by lack of depth data near the coastal area. For oceanic current model, GFS and NFS-MC CWB wind forecast were used as meteorology input. Initial fields and boundary condition are derived from HYCOM results. Nudging of salinity and temperature also were used to stabilize the model. Transport of Kuroshio of 2009 is about 17.0¡Ó3.2Sv. Maximum value is about 28.6Sv, occurred in summer. Minimum value is about 8.3Sv, occurred in winter.

Smart grid is an advanced electrical grid equipped with communication capability which is utilized to improve the efficiency, reliability, and sustainability of electricity services. Countries within Europe, North America and East Asia are undergoing a transformation from an antiquated infrastructure to the smart grid. However, some of problems arise due to the security and privacy issues of smart grid. Since smart meters and a grid operator can interact through its communication channel, there is a possibility that a hacker can hack into the system to steal information or even cut off the electricity service. Moreover, people are protesting and refusing to use smart meter since it enables the grid operator to perform frequent meter reading which unveils the customers¡¦ private energy usage information that could be abused. To cope with the privacy issue, we proposed an enhanced version of aggregation protocol from Garcia-Jacobs protocol where our protocol protects not only customers¡¦ energy consumption information but also the consumption information of a neighborhood. Furthermore, we proposed a novel privacy preserving billing protocol based on Priced Oblivious Transfer (POT) protocol which guarantees the grid operator to get the correct amount of money without knowing the individual energy consumption of the customers. Additionally, we also implement our proposed protocols.

In recent years, due to the issues of energy saving, the smart grid has become more important. AMI (Advanced Metering Infrastructure) is the basic of smart grid. However, AMI¡¦s network initialization usually cost a lot of time delay and energy waste because of many collisions due to the initialization in the high node density and variable network. In this paper, we proposed a Dynamic Contention Slot Initialization Protocol (DCSI Protocol) to reduce time delay and energy waste in the network initialization. At the beginning, all nodes in DCSI protocol are set in the receiving state. The proposed approach reduces not only collisions but also the communication failure due to the interference out of the transmission range. We divided time into time slots and then composed them to superframe. The first slot of superframe is designed for master node¡¦s broadcast, and other time slots are devised for other nodes to join in the network. Based on the previous superframe, nodes for the proposed protocol adjust the number of the contention slot by detecting collisions to adapt the high node density and variable network. The simulation results demonstrate superiority of DCSI protocol over flooding.

Smart Grid technology is likely to be implemented in various magnitudes across utilities in the near future. To accommodate these technologies significant changes will have to be incorporated in building design construction and planning. This research paper attempts to evaluate public utility executives’ plans to adopt smart grid technologies and to assess timing of smart grid impacts on future design and construction practices. Telephone survey was the data collection method used to collect information from executives at cooperative and municipal utilities. The study focuses on small and medium utilities with more than five thousand customers and fewer than one hundred thousand customers. A stratified random sampling approach was applied and sample results for fifty-nine survey responses were used to predict the timing of smart grid implementation and the timing of smart grid impacts on future design and construction practices. Results of this research indicate that design and construction professionals should already be developing knowledge and experience to accommodate smart grid impacts on the built environment.

Due to the advances of image system, most researches are developed on high-resolution system. However, the low-resolution system have an advantage over the high-resolution system on processing speed, saving space and power consumption. From the research of hexagonal grid, we know that from the view of microcosmic, the angle resolution and connection of hexagonal grid are better than rectangular grid, so images on hexagonal grid also have the better quality. Almost input and output systems of image device are on rectangular grid, so its technology and theory are developed on rectangular grid system. For a displaying system, fonts and images are the main elements. If we want to substitute hexagonal grid system for rectangular grid system, they will be primary factories that we consider. In this research based on rectangular grid system, we will apply the plane parametric curves and fill algorithm on hexagonal grid system, and probe into the displaying method of fonts and images on hexagonal grid system.

The design and implementation of a small-scale solar energy conversion system is presented in this thesis. The power circuit of the PV inverter is composed of a push-pull converter and a full-bridge inverter. The inverter converts DC power generated by photovoltaic cells into AC power. This design allows flexible arrangements of various solar cells modules. The systems can supply loads and feed excess power to the grid with unit power factor when operated under grid-connected mode, or supply loads exclusively by operating in the stand-alone mode. In grid-connected mode, it is controlled to follow the variations of solar panel insolation for achieving maximum power point tracking. A 150W prototype of the conversion system is implemented, simulation and experimental results are provided to verify the performance.

Images scaling is a very common capability on rectangular grid. Similarly, its development on hexagonal grid is very fundamental and necessary. In this paper, we developed new techniques to scale digital image with resampling and without resampling. From now on, we haven¡¦t seen any research about color displaying on hexagonal grid. It limited the application of hexagonal grid displaying device. In this paper, we developed a color displaying system on hexagonal grid by using symmetrical triangular frame.