基于特高频的开关柜局部放电检测与故障诊断

Partial Discharge Detection and Fault Diagnosis of Switch Cabinets Based on UHF

特高频具有灵敏度高、抗干扰能力强等优点,是开关柜局部放电检测领域最有效的检测方法之一。根据开关柜常见的故障类型设计并加工制作了尖尖、尖板、气隙、沿面4种局部放电模型,同时搭建了特高频局部放电检测平台,然后采用传统的统计特征进行参数提取,最后采用PSO-SVM算法进行故障诊断,并进行统计与时频域特征向量提取方式、BP神经网络故障诊断算法对比。结果表明,基于统计特征的PSO-SVM算法的识别率能达到98.5%,具有一定的工程参考价值。

Ultra high frequency(UHF)has the advantages of high sensitivity and strong anti-interference ability.It is one of the most effective detection methods in the field of partial discharge detection of switch cabinets.In this paper,four partial discharge models of tip,tip plate,air gap and surface were designed and manufactured according to the common fault types of switch cabinets,at the same time,the UHF partial discharge detection platform was built,and then the traditional statistical features were used for parameter extraction.Finally,the PSO-SVM algorithm was used for fault diagnosis,and the statistics was compared with the time-frequency domain feature vector extraction method and BP neural network fault diagnosis algorithm.The results show that the recognition rate of PSO-SVM algorithm based on statistical features can reach 98.5%,which has a certain engineering reference value.