摘要
基于油纸绝缘气隙放电模型,研究放电能量的变化规律。通过引入放电重复率,提出用每秒平均放电能量描述气隙放电的发展过程。引入放电发生的工频相位,构造基于放电能量的φ-W-n三维统计图谱,投影到放电能量-相位平面得到灰度图像。采用一种基于小波矩特征的局部放电灰度图像特征提取方法,得到灰度图像的全局特征和局部特征。根据提取到的不同放电时刻的小波矩特征值,采用模糊C-均值聚类的方法将整个放电过程划分为放电产生和振荡发展阶段、微弱放电阶段、放电爆发阶段以及放电预击穿阶段。
The variation law of discharge energy is researched based on the model of air-gap discharge in oil-paper insulation. The discharge repetition rate is introduced and the average discharge energy per second is proposed to describe the developing process of air-gap discharge. The phase of power frequency is introduced and a three-dimensional statistic graph φ-W-n is constructed based on the discharge energy,which is then projected to the discharge energy-phase plane to obtain the gray intensity image. A feature extraction method based on wavelet moment feature is adopted to extract the global and local features of the obtained gray intensity image. Based on the wavelet moment feature values extracted at different times,the whole discharge process is divided into four stages by the fuzzy C-means clustering method:the discharge generation and oscillation development stage,the weak discharge stage,the discharge outbreak stage and the pre-breakdown stage.
出处
《电力自动化设备》
EI
CSCD
北大核心
2014年第6期89-95,共7页
Electric Power Automation Equipment
基金
国家创新研究群体基金资助项目(51021005)
国家重点基础研究发展计划(973计划)项目(2009CB724506)~~
关键词
变压器
绝缘
油纸绝缘
气隙放电
放电能量
小波矩特征
模糊C-均值聚类
局部放电
electric transformers
insulation
oil-paper insulation
air-gap discharge
discharge energy
wavelet moment characteristics
fuzzy C-means clustering
partial discharges