Journal of the Korean Solar Energy Society ISSN:1598-6411(Print) 2508-3562(Online) 한국태양에너지학회 논문집

Considering the dangers of DC serial arcs in PV systems and the time when arcs are caused by electrical fire, we aimed to quickly detect DC serial arcs within 0.5 s. To quickly analyze and identify the frequency characteristics of arcs, fast Fourier transform (FFT) and discrete wavelet transform (DWT), which have been the most widely used frequency analysis methods in recent years, were compared and analyzed to implement an algorithm for detecting the most accurate and efficient DC arc. In this study, an experiment was conducted based on the UL 1699B arc test conditions. At this time, the arc energy does not exceed 750 J, and the arc must be detected within 2.5 s. In the overall process, components other than the arc were removed by applying a band pass filter to detect a specific frequency component of the arc. ADC was performed to convert the current value into a digital signal, and signal processing was performed using FFT and DWT using the filtered signal. Finally, after calculating the energy, which is a component of the frequency for each specific period, the threshold of each energy value was set to determine whether the arc exceeds a specific threshold.