CEEMDAN在变压器振动信号提取中的应用

有效提取变压器振动信息对于识别变压器运行状态具有重要的现实意义.本文提出一种基于自适应白噪声完整集成经验模态分解(CEEMDAN)的变压器振动信号分析方法.论文首先采用CEEMDAN将原始变压器振动信号分解成有限的固有模态分量(IMF),之后利用相关系数(CC)提取有效的IMF分量.最后,将提取出的有效IMF分量进行信...

Application of CEEMDAN in vibration signal extraction of transformer

It is of great practical significance to extract transformer vibration information effectively for the identification of transformer operation state. In this paper, a new method of transformer vibration signal analysis based on adaptive white noise integrated empirical mode decomposition (CEEMDAN) is proposed. Firstly, the original transformer vibration signal is decomposed into limited intrinsic mode function (IMF) by using CEEMDAN, and then the effective IMF component is extracted by calculating correlation coefficient (CC). The effective IMF components are reconstructed to obtain the denoised results. The simulated signal and on-site signal analysis show that the method can achieve signal denoising accurately and extract the effective information of transformer vibration signal successfully. It verifies the effectiveness and practicability of the method.