摘要
针对变压器有载分接开关机械故障诊断,引入一种基于经验模态分解EMD(empirical mode decomposi-tion)能量熵的诊断方法,以提取变压器有载分接开关振动信号特征并进行故障诊断。首先进行小波消噪,然后对信号进行经验模态分解,对得出的各阶固有模态函数求能量,最终计算得到信号的能量熵值。运用EMD能量熵作为特征参量,分析了触头正常和烧毁两种情况下分接开关切换时产生的振动信号,并与小波能量熵比较,研究结果表明,分接开关振动信号基于EMD能量熵的方法比基于小波能量熵的方法有效。
For the mechanical fault diagnosis of transformer on-load tap changer （OLTC）, a kind of energy entropy based on empirical mode decomposition （EMD） is introduced to extract characteristics of vibration signal and diagnose fault in OLTC. Firstly the noises of vibration signal are eliminated by wavelet, and then decom- pose it by EMD to get the intrinsic mode functions （IMFs）. After calculating the IMFs＂ energy, the EMD en- ergy entropy can be gotten. In this paper, EMD energy entropy is used as characteristic parameter to analysis vibration signal when OLTC is normal and burned-out. Comparison between EMD energy entropy and wavelet energy entropy showed that EMD energy entropy which used in vibration signal of OLTC is more effective than wavelet entropy.
出处
《电力系统及其自动化学报》
CSCD
北大核心
2012年第4期48-53,共6页
Proceedings of the CSU-EPSA
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
经验模态分解
小波分解
能量熵
on-load tap changer（OLTC）
vibration
empirical mode decomposition（EMD）
wavelet decompo-sition
energy entropy
