| Summary: | 碩士 === 國立臺灣科技大學 === 電機工程系 === 105 === The renewable energy is rising in recent years, and so the proportion of DC power in power system is gradually increased. DC arc fault protection is also becoming more important in system. The arcing will accompany with high temperature and sparking that is easy to cause serious fires. Several oversea examples of fires caused by arcing fault at PV system were reported. USA and Taiwan have announced regulations that PV systems shall include DC arc fault circuit protection devices. In this thesis, it is to establishe an experiment platform of DC arc fault. First of all, the experimental data are analyzed by the Fast Fourier Transforms (FFT) for the signal processing to capture characteristics of current. Thereafter, these extracted characteristics are applied to the fuzzy system and the learning vector quantization (LVQ), respectively. Therefore, the two detecting methods are designed. At last, two detecting methods are implemented by using the Field Programmable Gate Array (FPGA) preliminarily. The detecting results are compared to commercial photovoltaic arc fault detection (PV AFD). According to the experiment results, these detecting methods can correctly and effectively detect series arc fault. And in the case of power circuit growth, they have better ability than commercial AFDs.
|
|---|
