GIS (Gas Insulated Switchgear) internal partial discharge defect recognition method based on PRPS signals

• Main Authors: TIAN LIBIN, ZHU YUNJIA, JIN XIEJIE
• Format: Patent
• Language: Chinese; English
• Published: 13.11.2018
• Subjects: CALCULATING; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; MEASURING; MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; TESTING

The invention discloses a GIS (Gas Insulated Switchgear) internal partial discharge defect recognition method based on PRPS signals. The GIS internal partial discharge defect recognition method comprises the steps: obtaining a lot of partial discharge signals PRPS through a GIS built-in sensor; extracting a plurality of characteristics of a discharge capacity mean value, a discharge duality rate,initial discharge phase window difference, a discharge width ratio, a discharge phase mean value, discharge phase standard deviation, discharge phase skewness and the like; utilizing extreme gradientto improve a classification tress to achieve establishment of a PRPS defect type recognition model. According to the method disclosed by the invention, machine learning is utilized to perform discharge defect recognition on PRPS atlases obtained through an ultrahigh frequency method, a traditional time-consuming complex process that a GIS partial discharge defect recognition system is achieved byconverting PRPS to PRPD is