基于小波变换与SVD的水电机组振动信号特征提取研究

Research on Feature Extraction of Hydropower Units Vibration Signal Based on Wavelet Transform and SVD

提出一种基于小波变换与SVD相结合的方法用于提取水电机组振动故障特征。运用小波变换对已去噪处理的水电机组振动信号进行变换,变换得到信号各分支的小波分解系数,对各分支系数进行差值单支重构后,组成SVD的输入矩阵,提取奇异值得到特征向量。应用概率神经网络对提取的奇异值特征量进行效果分类。通过水电站机组实测数据验证表明该特征提取方法操作简单稳定,具有较高区分度与较好识别率,可以为水电机组状态故障诊断提供有效依据。

In this paper,a method based on wavelet transform and SVD is proposed to extract vibration fault characteristics of hydropower units.The wavelet transform was used to transform the de-noised vibration signal of hydropower unit.The wavelet decomposition coefficients of each branch of the signal are obtained by the transformation.After single branch reconstructing the differences of the branch coefficients,the SVD input matrix is formed and the singular value is extracted to obtain the feature vector.Probabilistic neural networks are used to classify the extracted singular value features.The test results show that the method is simple and stable,with higher division and better recognition rate,which can provide an effective basis for the diagnosis of hydropower unit fault.