基于小波系数区域相关性的关联维数在风机故障诊断中的应用

Application of correlation dimension in fan fault diagnosis based on wavelet coefficient region relativity

提出用小波系数区域相关性的滤波算法对原始数据进行降噪处理,用关联维数来定量描述风机的工作状态,进而对风机进行故障诊断。计算了风机在不同工作状态下降噪前后振动信号的关联维数。分析结果表明,该方法能有效的保留信号特征,有较好的降噪效果,且风机在不同工作状态下所对应的关联维数有明显不同,因此可以用关联维数区分风机的工作状态,以关联维数作为风机故障诊断的依据。

The method of fan fault diagnosis is studied. A threshold filtering algorithm based on region relativity of the wavelet coefficients is used to process the raw data and to reduce the noise of fan. The correlation dimension is employed to quantitatively reflect the working conditions. The correlation dimensions of vibration signals of fan before and after noise reducing under different working conditions are calculated using this method.The results show that this method is effective.