广东电力2018，Vol.31Issue(2)：119-124,6.DOI:10.3969/j.issn.1007-290X.2018.002.019
基于经验模态分解的变压器振动信号盲源分离
Blind Source Separation Technology of Transformer Vibration Signal Based on Empirical Mode Decomposition
摘要
Abstract
Transformer vibration signal contains a wealth of winding,core and other spare parts of the health status informa-tion.The real-time running vibration signal collected on the surface of the transformer is the aliasing signal of the core and winding vibration signal,The separation of the aliasing signal is more conducive to transformer fault diagnosis. A blind source separation method based on EMD decomposition is proposed,the signal is decomposed by EMD firstly,then extract the most relevant IMF components,finally,the fast ICA algorithm is used to solve the separation matrix.And the method is applied to the experimental simulation signal,which proves the feasibility of the method.Finally,the method is applied to the running transformer,the results show that the method can successfully separate the transformer core and winding vibra-tion signal,the effect is good.关键词
变压器/振动信号/盲源分离/经验模态分解Key words
transformer/vibration signal/blind source separation/empirical mode decomposition分类
信息技术与安全科学引用本文复制引用
谢荣斌,张丽娟,靳斌,李诗勇,刘波,赵莉华..基于经验模态分解的变压器振动信号盲源分离[J].广东电力,2018,31(2):119-124,6.基金项目
贵州电网有限责任公司科技项目(060100〔2016〕030301SY503) （060100〔2016〕030301SY503）
