基于Hilbert模量频谱分析的开关柜局部放电故障在线监测方法研究  

Research on Online Monitoring Method for Partial Discharge Faults in Switchgear Based on Hilbert Modulus Spectrum Analysi

作　　者：孙旭 程瑞剑 邱伊键 刘琦 熊剑 SUN Xu;CHENG Ruijian;QIU Yijian;LIU Qi;XIONG Jian(Jiangxi Key Laboratory of Advanced Copper-based Materials,Institute of Applied Physics,Jiangxi Academy of Sciences,330096,Nanchang,PRC;Institute of Energy Research,Jiangxi Academy of Sciences,330096,Nanchang PRC;Jiangxi Yuanfeng Electric Power Co.,Ltd.,330096,Nanchang PRC)

机构地区：[1]铜基新材料江西省重点实验室,江西省科学院应用物理研究所,南昌330096 [2]江西省科学院能源研究所,南昌330096 [3]江西源丰电力有限责任公司,南昌330096

出　　处：《江西科学》2024年第6期1220-1225,共6页Jiangxi Science

基　　金：江西省科学院省级科技计划包干制试点示范重点研发一般项目(2022YSBG22022)。

摘　　要：针对开关柜局部放电故障难以实时准确监测的问题,开展基于Hilbert模量频谱分析的在线监测方法研究。通过设计超高频信号前置处理,有效滤除噪声干扰,提升信号质量。采用Hilbert模量频谱分析技术,深入提取局放信号的特征参数,特别是低中频能量百分比等关键特征。结合特征参数,计算特征与各放电故障类型的马氏距离,实现对开关柜局放故障的在线监测与精准识别。通过对比实验证明,该方法能够显著提高故障检测的准确性,为开关柜的安全运行提供有力保障。In response to the issue of difficulty in real-time and accurate monitoring of partial discharge faults in switchgear,research on online monitoring methods based on Hilbert modulus spectrum analysis is carried out.By designing ultra-high frequency signal preprocessin g,noise interference is effectively filtered out to enhance signal quality.Utilizing Hilbert modulus spectrum analysis technology,the characteristic parameters of partial discharge signals are deeply extracted,especially key features such as the percentage of low and intermediate frequency energy.By combining feature parameters,the Mahalanobis distan ce between features and various types of discharge faults is calculated to achieve online monitoring and accurate identification of partial discharge faults in switchgear.Through comparative experiments, it has been proven that this method can significantlyimprove the accuracy of fault detection and provide a strong guarantee for the safe operationof switchgear.

关 键 词：Hilbert模量频谱分析 局部放电 在线监测 放电故障 开关柜 

分 类 号：TM507[电气工程—电器]