摘要
通过搭建球板电极的油纸绝缘实验平台,对单个局部放电波形的几何特征进行分析,构成最大幅值概率密度分布图、单个波形总时间概率密度分布图和最大幅值时间所占比例概率密度分布图。对每一个概率密度分布图提取平均值、方差、斜度和陡峭度4个特征量,然后利用K均值聚类算法进行聚类,对球板电极油纸绝缘的缺陷发展程度进行分类。结果表明:单个局部放电波形中各特征量的平均值与分散度随着油纸绝缘缺陷的发展具有明显不同,缺陷发展程度分为3个阶段,由此可以判断油纸绝缘的缺陷发展程度。
An oil-paper insulation experiment platform of ball-plate electrode was built and the geometric characteristics of single partial discharge waveform were analyzed to constitute the probability density distribution images of maximum amplitude, total time of single waveform, and ratio of maximum amplitude time. Four characteristic parameters including average value, variance, skewness, and steepness were extracted from each probability density distribution image, and then K-means clustering algorithm was used to cluster them so as to classify development degrees of oil-paper insulation defect. The results show that the average value and dispersion degree of each characteristic parameter in a single partial discharge waveform are obviously different with the development of oil-paper insulation defects, and the development degree of defect can be divided into three stages, which can be used to judge the development degree of oil-paper insulation defect.
作者
张鸿儒
史瑞楠
马帅
刘宾
李清泉
ZHANG Hongru;SHI Ruinan;MA Shuai;LIU Bin;LI Qingquan(Shandong Provincial Key Laboratory of UHV Transmission Technology and Equipment, School of Electrical Engineering, Shandong University, Jinan 250061, China;Maintenance Branch of State Grid Jiangsu Electric Power Company Limited, Nanjing 211106, China;Laiwu Power Supply Company of State Grid Shandong Electric Power Company, Laiwu 271100, China)
出处
《绝缘材料》
CAS
北大核心
2018年第12期53-58,共6页
Insulating Materials
基金
国家自然科学基金资助项目(51777117)
关键词
换流变压器
球板电极
单个波形
K均值聚类
缺陷发展程度
converter transformer
ball-plate electrode
single waveform
K-means clustering
development degree of defect