摘要
直流局部检测的意义重大,其数据分析也是很关键的。针对高压电容器直流局部放电测试中的离散数据和有限的数据量,本论文借助统计理论,探索性地选截尾均值进行数据前处理,然后根据统计量的物理意义,对前处理的数据进行峰度、偏度、相关度、变异度等参量提取,探索到不同性能的电容器其指纹参量具有一定的可比性,试尝找到同各类型的高压电容器在绝缘性能上的个体差异,这对以后的深入研究有很大的参考价值。
Partial discharge analysis under direct current voltage and analysis of PD data are very important.Since the data in PD test for HV capacitor are discrete and limited,the paper proposed trim-mean method for pretreatment by statistical theory. Kurtosis, Skewness, correlation coefficiet, and variance eoefficient are extracted from pretreatment according to statistical physics. The author find that different capability capacitor has different PD fingerprint and try to distinguish the respective insulating capability of the same types capacitors, which is of great importance for further research.
出处
《中国测试技术》
2006年第2期46-48,共3页
CHINA MEASUREMENT & TESTING TECHNOLOGY
基金
中国工程物理研究院电子工程研究所创新基金(S20030402)
关键词
局部放电
截尾均值
偏度
相关系数
变异系数
PD
Trimmean
Skewness
Correlation coefficient
Variance coefficient
