希尔伯特变换在电机故障诊断中的应用

HILBERT TRANSFORM AND ITS APPLICATION FOR DIAGNOSING FAULTS OF ELECTRICAL MACHINES

电机的许多故障信息是以调制的形式存在于所监测的电气信号及振动信号之中，如果借助于希尔伯特变换对这些信号进行解调处理，就能方便地获得故障特征信息，以确定电机所发生的故障类别。考虑到监测信号存在有各种噪声与干扰，须对监测信号作预处理以提高信噪比。文中介绍了希尔伯特变换的解调原理以及用ＦＦＴ子程序实现解调的步骤。通过对调制信号作谱分析，以其频谱中是否存在某种故障的特征频率分量来诊断有无该种故障。实验结果表明这种诊断方法是有效的。

The failure information of electric machines is contained in electrical and vibration signals monitored, which are often amplitude modulated. If Hilbert transform is used in processing the signal, the characterized fault information can be extracted from demodulated signal easily, and the type of fault can be determined. The preprocessing of the monitored signal is necessary for increasing signal to noise ratio. The principle of demodulation by Hilbert transform and its performing steps using FFT subroutine are introduced in this paper. The fault is diagnosed by extracting the fault feature component from the frequency spectrum of the modulated signal. The test results show that this diagnosis approach is effective.