变压器绕组多故障条件下的振动信号特征提取

Feature extraction for vibration signal of transformer winding with multiple faults

针对变压器绕组多种故障并发的工况,在分析变压器绕组振动机理的基础上,提出一种基于集合经验模式分解(EEMD)的振动信号特征提取方法。采用EEMD方法对变压器绕组振动信号进行分解得到各阶本征模函数(IMF),利用IMF能量和2范数构造特征矢量,将该特征矢量作为变压器绕组状态识别的判据。利用Fisher判别法对所提方法的有效性进行验证。试验结果表明,利用所提方法提取的各状态特征矢量区别明显,与快速傅里叶变换(FFT)方法相比,所提方法可准确识别出变压器绕组的混合故障状态。

The mechanism of transformer winding vibration is analyzed and a feature extraction method based on EEMD（Ensemble Empirical Mode Decomposition） is proposed for the vibration signal of the transformer winding with multiple faults. The vibration signal is decomposed by EEMD method to get the IMF ＇Intrinsic Mode Function） of each order and the feature vector is constructed with the IMF energy and 2-norm,which is then used as a criterion for the transformer winding state identification. Fisher discriminant is applied to verify the effectiveness of the proposed method. Experimental results show that, the feature vector extracted by the proposed method is significantly different among different transformer winding states and, compared with the FFT（Fast Fourier Transformation） method,the proposed method can properly identify the muhi-fauh state of transformer winding.