Two-stage UHV GIL Fault Location Based on Ultrasonic Sensor Network

Abstract

The failure of a long-distance Ultra-high voltage (UHV) Gas Insulated Transmission Line (GIL) during operation will cause great security risks to power transportation. How to locate GIL faults quickly and accurately and further carry out related operation and maintenance work is of great significance to the stable operation of the power grid. Based on this, this paper proposes a two-stage fault location method for UHV GIL based on an ultrasonic sensor network. In this method, ultrasonic sensors are configured in the shell of GIL to form an ultrasonic sensor network and internal faults are located by analyzing the signals collected from the network. The method is divided into two stages: 1) according to the collected signal data, the two sensors closest to the fault point are preliminarily identified, and then the fault point is roughly located, and obtain the data collected by the sensors; 2) we analyze sensor signal data, train and establish a network model based on recurrent neural network RNN for accurate fault location. The verification experiment shows that this method has fast location speed and high location accuracy, and can be effectively applied for quick maintenance, safe operation, and maintenance of GIL.