基于MMSE和谱峭度的滚动轴承故障诊断方法

Fault diagnosis method for rolling bearings based on MMSE and spectral kurtosis

共振解调作为滚动轴承常用的故障诊断方法,存在带通滤波器参数难以确定的缺点,针对此缺点以及滚动轴承早期故障信号信噪比低的问题,结合最小均方误差估计方法(MMSE)和基于谱峭度的共振解调方法,并将其应用于滚动轴承早期微弱故障诊断中。首先用MMSE方法抑制白噪声来提高信噪比,然后利用谱峭度自适应确定最优带通滤波器参数,最后对带通滤波后的信号进行能量算子解调谱分析,得出诊断结果。数字仿真信号和实验信号验证了该方法的有效性。

The demodulated resonance technology, as a common method for fault diagnosis of rolling bearing, has the deficiency of difficulty to choose the parameters of band-pass filter. In order to solve this problem and apply it in the occasion of low signal-to-noise ratio of original signal, a method of rolling bearing fault diagnosis was presented based on the combination of MMSE method and spectral kurtosis. In the method, white noise was restrained to increase signal-tonoise ratio, then the best resonance frequency band parameters were extracted by spectral kurtosis automatically, and then the results were achieved by the energy operator demodulation spectrum analysis of band-pass filtered signal. The results of numerical simulation and the data of an experiment show that the proposed method is effective for diagnosing the fault of rolling bearing.