摘要
针对基于经验模态分解(EMD)的希尔伯特黄变换(HHT)在变压器有载分接开关(OLTC)振动信号故障检测方面的模态混叠问题,提出了集合经验模态分解(EEMD)方法,有效解决了EMD模态混叠的缺点。在EEMD的基础上对振动信号进行HHT,提取信号的边际谱,实现对触头松动故障的检测。仿真与实际试验表明,将该方法应用于分接开关振动信号的分析,可有效地提取故障特征量,有利于对触头故障的诊断。
Aiming at the mode mixing problem,which is exists in the vibration signal fault detection for transformer on-load tap changer(OLTC) using Hilbert-Huang transform(HHT) based on empirical mode decomposition(EMD),the ensemble empirical mode decomposition(EEMD) method was put forward;and the mode mixing problem of EMD had been proved effectively.Based on the EEDM,the vibration signal can be Hilbert-Huang transformed;the marginal spectrum can be drawn from the HHT result;then the detection of the contacts loosening fault can be realized.The simulation and practical experiment show that to use this method in analysis of vibration signal of the tap changer can draw the values of the fault characteristic parameters effectively;and this is useful for contacts fault diagnosis.
出处
《华电技术》
CAS
2012年第1期12-15,76,共4页
HUADIAN TECHNOLOGY
基金
江苏省电力公司重点科技项目(J2008039)
关键词
有载分接开关
振动信号
故障诊断
集合经验模态分解
希尔伯特黄变换
边际谱
on-load tap changer
vibration signal
fault diagnosis
ensemble empirical mode decomposition(EEMD)
HilbertHuang transform(HHT)
marginal spectrum