Detection method of low voltage series DC arc based on the pattern matching algorithm

The paper presents a method of the series arc fault for DC systems. The aim of the study is to find some current characteristics, which can be significant for the purpose of arc detection. The validity of the proposed method is testified by analyzing the arc fault current signal. The low voltage series DC arc study is carried out on the currents waveforms acquired during some tests on electric appliances. A Fast Fourier Transformation(FFT) is applied to decompose the monitored current signals into a series of detailed arcing features. Dynamic time warping (DTW) is used to measure the input and the known arc model. The pattern matching algorithm is then used to detect and classify the arc fault. As a result, the method is effective in detecting and classifying the arc fault, and easy to be realized in the practical cases.