基于谱峭度和Teager能量算子的轴承故障特征增强

Bearing Fault Feature Enhancement Method Based on Spectral Kurtosis and Teager Energy Operator

共振解调法的难点在于带通滤波器的确定,谱峭度可根据信号特征寻找最优滤波器参数,很好地解决以上问题。然而谱峭度在对低信噪比数据进行处理时,滤波后的信号往往残留较大带内噪声,极大地影响了后续故障诊断的准确性。针对该问题,提出利用Teager能量算子追踪SK滤波信号的系统总能量,从信号能量的角度消除带内噪声,二次增强隐藏于噪声中的故障冲击特征,最后通过包络谱分析获得诊断结果。应用轴承故障仿真数据、实验室内圈和外圈故障数据验证了本方法的有效性。

The difficulty of resonance demodulation method is to determine the parameters of band-pass filters.Spectral kurtosis(SK) can resolve this conundrum by searching for the optimal parameters of the band-pass filters according to the characteristics of signals.However,in the case of strong background noise,SK is inadequate to extract local fault induced cyclic impact from bearing vibration signals due to the in-band noise.Aimed at this issue,a novel fault diagnosis method for rolling bearings is proposed based on spectral kurtosis and Teager energy operator.The Teager energy operator is employed to calculate the energy of signals after the SK,which can suppress the in-band noises and further enhance the cyclic impact feature hidden in vibrations of faulty bearings after filtering.The frequency spectrum of the signal energy is then given to determine the health condition of bearings.The effectiveness of the method is examined by using both simulative and experimental data.