A comparative analysis of characteristics of different partial discharge signals detected by F-P optical fiber sensor in the transformer

Unlike other kinds of optical fiber testing methods, Fabry-Perot (F-P) optical fiber ultrasonic sensing technology is to detect the ultrasonic generated in the process of partial discharge (PD) in the oil. As F-P optical fiber sensor is highly sensible, resistant to interference and high-voltage potential, simple in structure and can be installed in the transformer, it guarantees an accurate and effective monitoring of partial discharge in the transformers. In order to identify the differences of signals´ characteristics detected by F-P optical fiber sensors in different types of faults, this paper builds four typical partial discharge models, namely suspended discharge, point discharge, surface discharge and oil gap discharge. As different forms of partial discharge have different time-frequency characteristics, equivalent time-frequency method is chosen to compare the characteristics of the signals and different time-frequency diagrams are given. Patterns of partial discharge are distinguished through the comparative analysis of these time-frequency diagrams.