期刊文献+

基于EMD优化双谱的开关柜局部放电特高频信号时延计算方法研究 被引量:5

Time Delay Algorithm for UHF Partial Discharge Signals in Switchgear by EMD Based Improved Bispectrum Estimation Approach
下载PDF
导出
摘要 局部放电特高频信号的时延估计是对开关柜放电缺陷进行精确定位的基础,提出了一种基于经验模分解优化双谱的开关柜局部放电特高频信号时延计算方法。首先根据实验室的开关柜尺寸,采用Xfdtd软件立开关柜的模型,对其进行特高频电磁波传播仿真计算,在不同检测点获取局部放电特高频信号。然后在实室搭建开关柜局部放电特高频检测试验平台,在不同位置获取局部放电特高频信号。最后针对局部放电特高仿真和实测信号,采用所提的改进双谱法以及现有3种常用方法对不同信号的时延进行计算。结果表明所提法更适用于实际开关柜局部放电缺陷的精确定位。 The time delay estimation of partial discharge(PD)ultra-high-frequency(UHF)signals is the basis for the accurate positioning of discharge defect in switchgear.This paper presents an improved bispectrum estimation approach based on empirical mode decomposition(EMD)to calculate the time delay of the PD UHF signals in the switchgear.Firstly,a simulation model of switch cabinet is established by Xfdtd software based on the size of the switchgear in laboratory,a calculation about the PD UHF electromagnetic wave propagation is made,and the PD UHF signals are obtained at different detection points.Then UHF test platform for the PD in the switchgear is built in the laboratory,and the PD UHF signals are obtained at different locations.Finally,the improved bispectrum method and three existing methods are used to calculate the time delay of PD UHF signals from simulation and actual measurement.The results show that the EMD based improved bispectrum can provide the precise positioning of the discharge defect in the switchgear.
作者 何志满 王剑飞 蒋新川 江天炎 HE Zhiman;WANG Jianfei;JIANG Xinchuan;JIANG Tianyan(State Grid Wanzhou Power Supply Company,Chongqing 404100,China;State Key Laboratory of Power Transmission Equipment&System Security and New Technology,Chongqing University,Chongqing 00030,China)
出处 《智慧电力》 北大核心 2018年第11期59-64,共6页 Smart Power
基金 国家自然科学资助项目(51507017)~~
关键词 开关柜 局部放电 时延估计 经验模态分解 双谱法 switchgear partial discharge time delay estimate EMD bispectrum method
  • 相关文献

参考文献12

二级参考文献159

共引文献154

同被引文献51

引证文献5

二级引证文献43

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部