基于FrFT滤波和LMS降噪的变转速滚动轴承故障诊断

变速工况下的机械故障诊断逐渐成为旋转机械监控领域的一个热门课题，在变转速下故障更容易发生且伴随更大的噪声，而相应的降噪问题目前却没有可靠的解决方法。因此提出一种基于分数阶傅里叶变换（FrFT）滤波和最小均方算法（LMS）降噪的故障诊断方法，对变转速工况下轴承振动信号进行降噪，进而提取非平稳故障特征。首先，同时获得滚动轴承振动加速度信号和转速信号；然后对Hilbert解调后的振动信号进行峰值搜索FrFT，按照搜索得到的最佳阶次和分数阶域聚集位置进行FrFT滤波；再将FrFT滤波得到的信号作为参考信号，原包络信号作为输入信号，进行LMS自适应降噪；最后对降噪后的信号按照转速重采样进行阶次分析，将包络阶次谱中的突出特征与故障特征阶次对比，判断故障。该方法可成功应用于变转速工况下滚动轴承的试验数据处理，证明了方法的有效性。

Bearing fault diagnosis under varying speed conditions based on fractional Fourier transform(FrFT)filtering and least mean squares(LMS)noise reduction

Diagnosis of mechanical faults under varying speed conditions has gradually become a hot topic in the field of rotating machinery monitoring. Faults are more likely to occur with varying speeds, and are accompanied by greater noise, but there is currently no reliable solution for reducing such noise. In order to address this problem, a fault diagnosis method based on fractional Fourier transform (FrFT) filtering and least mean squares (LMS) noise reduction is proposed as a way of reducing the noise of bearing vibration signals under varying speed conditions, and non-stationary fault features can then be extracted. In the first step, the vibration acceleration signal and the speed pulse signal are obtained simultaneously. Next, the peak search FrFT is performed on the demodulated signal using the Hilbert transform, and FrFT filtering is employed to search for the best FrFT order and aggregation position. Then the signal obtained by FrFT filtering is set as the reference signal, and the original envelope signal is set as the input signal and LMS noise reduction carried out. Finally, order analysis is performed on the signal after noise reduction, and the order in the envelope order spectrum is compared with the fault characteristic order as a means to diagnose the fault.