摘要
提出基于特征矩阵和决策树的GIS电压互感器击穿放电故障识别方法。提取GIS电压互感器击穿放电故障的气体特征并构建特征矩阵,基于ID3算法生成决策树来对特征矩阵进行分类,从而识别不同的GIS电压互感器击穿放电故障类型。在实验中,应用所提出的方法获得了95.8%的GAS-VT击穿放电故障识别精度。
The present work made a preliminary attempt to explore the use of feature matrix and decision tree in the identification of disruptive discharge faults with respect to GIS voltage transformers.The method was designed to extract the gas characteristics of GIS voltage transformer encountering disruptive discharge fault for the construction of feature matrices,and to generate a decision tree using the ID3 algorithm for the classification of the feature matrices,so as to realize fault identification.The proposed method achieved in the experiment an identification accuracy of 95.8%for the GAS-VT disruptive discharge fault.
作者
李淑梅
满林林
孙毅
LI Shumei;MAN Linlin;SUN Yi(SDEE Hitachi High-Voltage Switchgear Co.,Ltd.,Jinan 250001,China)
出处
《电工技术》
2024年第20期189-191,194,共4页
Electric Engineering
关键词
特征矩阵
决策树
GIS电压互感器
击穿放电
故障识别
feature matrix
decision tree
GIS voltage transformer
disruptive discharge
fault identification