A Simulation-Based Approach to Design Flexible Hybrid Power Systems for Standalone Applications

Autor: Meng-HanLin, 林孟翰
Rok vydání: 2016
Druh dokumentu: 學位論文 ; thesis
Popis: 104
Integration of more than one renewable energy source for power generation is a research issue that has attracted considerable attention in recent years. Although the photovoltaic (PV) cell, the wind turbine (WT) and the fuel cell (FC) are all viable alternatives to the traditional steam and gas turbines, each is hampered by its own shortcomings. The PV cell alone is clearly not suitable for practical applications because of the intermittent nature of solar irradiation, while the power output of a WT unit is often unstable due to uncertain and uncontrollable weather conditions. The FC unit is fast to respond to demand changes but can be quite expensive. Finally, note that complementing a single power-generating unit with a large enough battery may not be an ideal solution either. This is due to the high investment cost of battery and its questionable durability in the long term. Thus, it is reasonable to expect that a properly configured hybrid system can provide uninterrupted off-grid power to meet the demand of a given community at any time. The goal of this study is to develop a systematic design methodology for configuring flexible hybrid power systems in standalone applications. For actual operation in a realistic environment, such a process must be fully operable despite random fluctuations in energy supplies and power demands. The authors therefore proposed a simulation based optimization strategy to compute two alternative measures, i.e., the dynamic and temporal flexibility indices, for quantitatively evaluating any given design. In order to demonstrate the usefulness of these criteria, the authors synthesized a large number of photovoltaic-fuel cell-wind turbine (PVFCWT) systems for the Annan Campus of National Cheng Kung University in Tainan, Taiwan, and then identified the best design(s) according to their costs and flexibility measures. In addition, the authors also developed a MATLAB/Simulink simulation code to validate these design decisions.
Databáze: Networked Digital Library of Theses & Dissertations