Application of signal processing and analysis in detecting single line-to-ground (SLG) fault location in high-impedance grounded distribution network

In high-impedance grounded distribution networks which are common in many countries, single line-to-ground (SLG) fault diagnosis is difficult due to weak fault characteristics, noise interference, three-phase unbalanced operation's influence and so on. Upon thorough analysis on the space-time characteristics of SLG fault, this study applies signal processing and analysis techniques to do SLG fault location detection. First, finite impulse response (FIR) filter is applied to effectively strain out interference factors and extract pure fault characteristics for each line. Afterwards, hierarchical clustering is used to classify lines based on the extracted fault characteristics. According to the clustering result, the fault line could be determined. If there are monitoring terminals distributed in the fault line, the fault section could also be further identified. Simulation shows that the proposed method works well under various fault conditions and has a higher reliability compared with wavelet analysis. Besides, this method makes full use of the existing hardware conditions in the grid and the cost is low, hence it has high practical value.