Online Monitoring and Identification of MV Distribution Line Fault
Zhang Bihui1, Yang Fan2, Liu Jian2, Guo Moufa1
1. College of Electrical Engineering and Automation, Fuzhou University, Fuzhou 350116; 2. State Grid Fujian Nanping County Electric Power Supply Co., Ltd, Nanping, Fujian 353000
Abstract：An online monitoring and identification system for medium voltage (MV) distribution line fault is developed to avoid the difficulty to exclude fault caused by long distance and wide distribution of MV distribution line. The system consists of hierarchical monitoring devices and PC software. The communication architecture is based on a communication method composed of RS-485, GPRS and Ethernet. Monitoring devices acquire and upload the fault waveform of line. PC software displays fault waveform and recognition result obtained by calling the fault type identification algorithm based on singular value decomposition (SVD) and multilevel support vector machine (SVM). The experimental results show that the system is stable and effective to monitor and identify line faults. In addition, the accuracy rate of proposed algorithm can reach 94%.
2017, Vol. 18 
