No description available.

With environmental and policy pressure to move towards cleaner fuel sources, wind energy is a proven technology that can be successfully implemented at the utility-scale and provide clean energy to the grid. Wind energy consists of many distributed wind turbines that are paralleled and connected to inject power to one location on the transmission grid. There are real power losses and reactive power drops that must be taken into consideration for these projects for plant performance and compliance. The better the performance of each new and operating wind farm installed, the more efficiently the grid operates as well as the less greenhouse gases generated. This thesis will first review the creation of an Excel tool to perform a load flow study given inputs for a wind farm using Newton-Raphson algorithms. Next, the results of the load flow analysis will be compared to an actual operating wind farm located in Texas to review the accuracy of the scenarios. Finally, alternative design and operating states for the wind farm are proposed and cases are simulated to review the impact on wind farm energy generation and reactive power provided to the grid. Finally, preferred improvements for future design and operational considerations are provided along with future areas of research and development. / Master of Science / This thesis is focused on improvements for wind farm design and operation to help wind farms generate more clean power to the grid. The thesis involves the creation of an Excel tool which can be used to complete required grid studies for real and reactive power flows within the wind farm to the point of connection with the transmission system. This analysis helps inform the wind farm design and operation to be more effective and operate more efficiently. An operating wind farm in Texas is explained and depicted for an understanding of how utility-scale wind farms are set up. Additionally, a year of data from an operating wind farm is used to compare the Excel load flow tool to actual data and confirm it's accuracy. Alternate methods this plant could have been designed and operated are evaluated using the new tool and actual operating conditions from the plant for the year under analysis are performed to better understand and quantify possible improvements for wind farms. This thesis is less focused on the wind turbine generator (WTG) construction and operation of a single unit, but rather focused on the output from the WTG and the impact on an entire system containing many of these distributed generators and their operation to provide energy to the grid.

Master of Science / Department of Electrical and Computer Engineering / Anil Pahwa / Warren N. White / This thesis focuses on estimating the power loss from power panelboards by means of power loss models. The model is intended to be used by HVAC engineers to help estimate building heat loss. While McDonald & Hickok (1985) did not report power losses for power panelboards, Rubin (1979) did. These publications present the power losses of electrical devices at rated loads in tables. In this thesis, the models for electrical devices are created and used, instead of tables, to estimate power losses. The use of curve fit models presents a convenience in calculation of power losses. Breaker, fusible switch, and motor starter power losses presented by McDonald & Hickok (1985) and Rubin (1979) were updated using manufacturer published data, technical papers, industrial standards, and test samples. Test, manufacturer, and analytical model data are collected and power loss curve fit models are created for breakers, fusible switches, motor starters, and bus bars with enclosures. The panelboard power loss is calculated as the sum of partial power losses of the component electrical equipment, i.e. breakers, fusible switches, motor starters, and bus bars with enclosures used in power panelboards. A power loss model for main breaker and fusible switch power panelboards are created based on the sum of breaker, fusible switch, motor starter, and bus bars with enclosure power loss models. The main breaker and fusible switch power panelboard power loss models are used in a heat loss example. It is shown that power panelboard power losses can be significantly overestimated when calculated with one of the methods currently used (Rubin, 1979). This can result in erroneous sizing of HVAC equipment.

The Fontan repair is a three-stage palliative surgical procedure for single ventricle congenital heart diseases, ultimately resulting in the right heart bypass. This is accomplished by routing the systemic venous return directly to the lungs. Although this procedure reduces the mortality rate, its long-term outcome is still considered far from optimal. Over the years several modifications have been suggested, ultimately leading to the total cavopulmonary connection (TCPC), which is the current procedure of choice. A better understanding of the hemodynamics in the TCPC is critical for further optimization of the TCPC design and surgical planning, which may lead to improved surgical outcome. Recent experimental and numerical studies have focused on characterizing the fluid dynamics of the TCPC but to date no study has attempted to relate the geometry of the TCPC anatomies with their hemodynamic parameters. The present study therefore proposes to quantify the complex geometrical characteristics of patient-specific TCPC anatomies and correlate these characteristics with their hemodynamic efficiency. A technique using skeletonization approach is thus developed to achieve this goal. The centerline approximation of the TCPC geometry is used to extract main geometric parameters such as vessel area, curvature and offset. The developed methodology is then applied to characterize the shape of various TCPC templates including extra-cardiac (EC) and intra-atrial (IA) TCPCs, TCPCs with bilateral Superior Vena Cavae and geometries before the third stage. The obtained geometric parameters are then related to the TCPC hemodynamics, particularly to the power loss.

The purpose of this study is to model an operational wind farm in DIgSILENT PowerFactory® using manufactures specifications and investigate the active power energy losses. The model is tested with the operational recorded data from the wind farm and is validated. Meeting the increased demand for renewably generated electricity drives the growth in wind energy which in turn gradually decrease the suitable locations to construct wind farms. On the other hand, the market forces persuade the wind farm developers to maximize the return on investments. Therefore, it is imperative to have optimized wind farm designs as well as accurate financial prognosis. Accurate models are important to estimate the wind farm characteristics while realistic loss estimations are needed for precise financial forecasting. One of the major components of the wind farm is the internal collector network through which the generated electricity is fed to the national electricity grid. With this study, modeling of internal collector network of an operational wind farm is carried out and the capability of the wind farm to maintain the stipulated voltage levels at the point of common coupling is examined together with the amount of cable loading. In addition, the active power energy losses, the impact of the internal collector network arrangement to the active power energy losses are investigated and a proposal to reduce the cost of cable laying for 4 cable sections is proposed for future wind farm designs. The findings of this report show that the internal collector network of the considered wind farm can maintain required voltage levels at the medium voltage busses of the network for different grid voltage levels. The operational data of power loss of the cables of the network match with the simulated results but the total loss does not. The reasons behind this mismatch could be the limited amount of operational data and measurement errors. Further analysis and comparison are suggested with larger sets of data together with the respective list of events to increase number of data points for the simulation.

Evolutions in integrated circuit technology require the use of a high-frequency synchronous buck converter in order to achieve low cost, low profile, fast transient response and high power density. However, high frequency operation leads to increased power MOSFET switching losses. Optimization of the MOSFETs plays an important role in improving converter performance. This dissertation focuses on revealing the power loss mechanism of power MOSFETs and the relationship between power MOSFET structure and its power loss. The analytical device model, combined with circuit modeling, cannot reveal the relationship between device structure and its power loss due to the highly non-linear characteristics of power MOSFETs. A physically-based mixed device/circuit modeling approach is used to investigate the power losses of the MOSFETs under different operating conditions. The physically based device model, combined with SPICE-like circuit simulation, provides an expeditious and inexpensive way of evaluating and optimizing circuit and device concepts. Unlike analytical or other SPICE models of power MOSFETs, the numerical device model, relying little on approximations or simplifications, faithfully represents the behavior of realistic power MOSFETs. The impact of power MOSFET parameters on efficiency of synchronous buck converters, such as gate charge, on resistance, reverse recovery, is studied in detail in this thesis. The results provide a good indication on how to optimize power MOSFETs used in VRMs. The synchronous rectifier plays an important role in determining the performance of the synchronous buck converter. The reverse recovery of its body diode and the Cdv/dt induced false trigger-on are two major mechanisms that impact SyncFET's performance. This thesis gives a detailed analysis of the SyncFET operation mechanism and provides several techniques to reduce its body-diode influence and suppress its false Cdv/dt trigger-n. This thesis also investigates the influence of several circuit level parameters on the efficiency of the synchronous buck converter, such as input voltage, circuit parasitic inductance, and gate resistance to provide further optimization of synchronous buck converter design.

This thesis proposes novel methods and comprehensive analysis for power loss calculation, DC-link current and voltage ripple estimation, and bus bar design in two-level three-phase voltage source inverters (VSIs). A novel method of MOSFET voltage rise- and fall-time estimations for the switching power loss calculation is developed. The estimation accuracy is significantly improved by the proposed method. In order to provide a reference for thermal management design, inverter power loss analysis is presented. Using the parameters obtained from the semiconductor device datasheets and inverter operating conditions, power loss calculations of three types of devices, namely IGBT, MOSFET, and diode, are discussed. The conduction power loss calculations for these three devices are straightforward; and, the switching power loss of IGBTs and diodes can be obtained from the energy losses given by datasheets. However, many MOSFET datasheets do not provide the switching energy losses directly. Therefore, to acquire MOSFET switching energy losses, switching transient times must be estimated as accurately as possible. The impacts of inverter anti-parallel diode reverse recovery on the DC-link current and voltage ripples are investigated. According to the analysis, the impact of diode reverse recovery on the voltage ripple is negligible, while the RMS value of current ripple is influenced by both diode reverse recovery and inverter switching frequency. A novel method is developed to calculate the ripple current RMS value and the estimation accuracy is significantly improved. Depending on the calculated current and voltage ripples, DC-link capacitor selection is introduced. Generally speaking, failures in the DC-link capacitors take place more frequently than the failures in other parts of the inverter system, and plenty of research has been focusing on minimizing the required DC-link capacitance. As a result, the accurate estimations of DC-link current and voltage ripples are vital in the optimization methods. In addition, with the accurate estimations, the over-design in the DC-link capacitance could be reduced. Finally, the design of a practical bus bar is presented. The DC current distribution is aff ected by the numbers and locations of the DC input tabs, while the AC current distribution is influenced by the numbers and locations of the installation holes for DC-link capacitors and semiconductor devices. Furthermore, parasitic parameters of the bus bar, especially the stray inductance and voltage spikes caused by this inductance during switching turn-o transients, are also discussed from the angle of the design rules and correlation between the parameters and bus bar geometry structure. In the end, a bus bar is designed with balanced current distribution and low stray inductance. / Thesis / Doctor of Philosophy (PhD)

Due to increasing penetration of solar power generation, distribution grids are facing a number of challenges. Frequent reverse active power flows can result in rapid fluctuations in voltage magnitudes. However, with the revised IEEE 1547 standard, smart inverters can actively control their reactive power injection to minimize voltage deviations and power losses in the grid. Reactive power control and globally optimal inverter coordination in real-time is computationally and communication-wise demanding, whereas the local Volt-VAR or Watt-VAR control rules are subpar for enhanced grid services. This thesis uses machine learning tools and poses reactive power control as a kernel-based regression task to learn policies and evaluate the reactive power injections in real-time. This novel approach performs inverter coordination through non-linear control policies centrally designed by the operator on a slower timescale using anticipated scenarios for load and generation. In real-time, the inverters feed locally and/or globally collected grid data to the customized control rules. The developed models are highly adjustable to the available computation and communication resources. The developed control scheme is tested on the IEEE 123-bus system and is seen to efficiently minimize losses and regulate voltage within the permissible limits. / Master of Science / The increasing integration of solar photovoltaic (PV) systems poses both opportunities and technical challenges for the electrical distribution grid. Although PV systems provide more power to the grid but, can also lead to problems in the operation of the grid like overvoltages and voltage fluctuations. These variations can lead to overheating and burning of electrical devices and equipment malfunction. Since the solar generation is highly dependent on weather and geographical location, they are uncertain in their output. The uncertainity in the solar irradiance can not be handled with the existing voltage control devices as they need to operate more frequently than usual which can cause recurring maintenance needs for these devices. Thus, to make solar PV more flexible and grid-friendly, smart inverters are being developed. Smart inverters have the capability of advanced sensing, communication, and controllability which can be utilized for voltage control. The research discusses how the inverters can be used to improve the grid profile by providing reactive power support to reduce the power losses and maintain voltages in their limits for a safer operation.

The aim of this thesis is to design a kilowatt three-phase step-down rotary transformer for a permanent magnet DC motor. The permanent magnet DC motor has an on-rotor drive system, and therefore requiring a power supply that can transfer power to its drive unit without mechanical contact. The rotary transformer has a detached magnetic coupling structure that qualifies it as a potential method for the wireless power transfer. This thesis studies the rotary transformer as a static device, focusing on its core loss. By using a transient finite element analysis of COMSOL Multiphysics and an iron loss prediction model, the rotary transformer was optimized in terms of efficiency and power density for the on-rotor drive system through proper material selection and geometry exploration. After this, a mechanical design, which based on a literature review of the influences of manufacturing processes on electrical steels, was proposed for realizing the core fabrication and the rotary transformer assembly. The results show that the rotary transformer can step down 400 V/50 Hz three-phase voltage to 13.15V in a Delta-wye connection and output 1.17kW power over an air-gap of 0.3mm with 95.94% overall efficiency. The proposed mechanical design enables the transformer to minimize the core loss and the manufacturing cost. Without using resonant inductive coupling, this transformer design simplifies the power supply for the motor, thereby decreasing the motor manufacturing and maintenance cost.

A reconfiguração de redes de distribuição de energia elétrica consiste em alterar sua topologia por meio de manobras de chaves nos circuitos primários. Trata-se de um problema de otimização combinatória, onde normalmente os objetivos são a minimização de perdas ativas e/ou número de manobras realizadas, atendendo as restrições como isolamento de faltas, balanceamento de cargas entre os alimentadores e melhoria dos níveis de tensão. As dificuldades na modelagem e na resolução exata de problemas envolvendo a reconfiguração de redes de distribuição advém do tamanho dos sistemas reais, representados por um número elevado de chaves e alimentadores e ainda pela natureza combinatorial do problema. Para tratar essas questões, diversas modelagens e técnicas computacionais têm sido desenvolvidas, em particular heurísticas de melhoramento que através de uma solução factível, otimiza os resultados reduzindo o espaço de busca, até encontrar uma nova solução com melhor função objetivo. Neste sentido, são propostas duas formulações matemáticas descrevendo novas restrições a fim de melhorar a descrição do problema. A primeira, uma formulação mais simplificada, considera apenas a parte ativa das instâncias; na segunda um modelo completo é descrito otimizando parte das restrições do primeiro modelo e considerando também a parte reativa das instâncias. Duas heurísticas também são adaptadas pela primeira vez para o problema de reconfiguração de redes, pois a heurística de melhoramento Fix-and-Optmize é configurada de duas formas diferentes, determinando seus principais parâmetros através de uma análise de sensibilidade. Os resultados dos dois modelos propostos e também das heurísticas adaptadas para 13 sistemas de referência são descritos e comparados com outros métodos da literatura. Para verificar a eficiência e robustez dos métodos e heurísticas desenvolvidos, replicações são propostas de dois sistemas de referência, 9 replicações do sistema de 72 barras e 4 replicações do sistema de 10560 barras. Seus resultados bem como o desempenho dos métodos são descritos e avaliados. / A reconfiguration of electricity distribution networks consists in altering a topology of the networks by means of key maneuvers in the primary circuits. It is a problem of combinatorial optimization, where the objectives are a minimization of active losses and/or number of maneuvers performed, taking into account constraints such as fault isolation, load balance between feeders and improvement of voltage levels. As difficulties in modeling and in the exact resolution of problems involving a reconfiguration of distribution networks come from the size of the real systems, represented by a large number of switches and feeders, and also by the combinatorial nature of the problem. To address these issues, several models and computational techniques have been developed, in particular heuristics of improvement that through a feasible solution, improves results by reducing the search space, until finding a new solution with better objective function. In this sense, in this thesis it is proposed of two mathematical formulations describing new constraints in order to improve a description of the problem. A first, simpler formulation considers only a active part of the instances, in the second a complete model is described optimizing some restrictions of the first model and also considering the reactive part of the instances. Two heuristics are also first adapted to the network reconfiguration problem. The Fix-and-Optmize enhancement heuristic is configured in two different ways, determining its key parameters through a sensitivity analysis. The results of the two proposed models and also of the heuristics adapted for 13 reference systems are described and compared with other methods of the literature. To verify the efficiency and robustness of the developed methods and heuristics, replications are proposed for two reference systems, 9 replications of the 72 bus system and 4 replications of the 10560 bus system. Its results as well as the performance of the methods are described and evaluated.