阶次跟踪能量算子与奇异值分解结合的滚动轴承故障诊断

Fault Diagnosis for Rolling Bearings Combining Order Tracking Energy Operator with Singular Value Decomposition

针对滚动轴承变速过程中振动信号非平稳,早期故障特征微弱的特点,将阶次跟踪与能量算子(EO)相结合,提出阶次跟踪能量算子(OTEO)与奇异值分解(SVD)结合的轴承故障诊断方法。首先,对时域振动信号进行SVD以降低噪声干扰;然后,对降噪后的信号进行OTEO解调分析,即对信号先进行EO解调再对包络信号进行阶次跟踪;最后,采用Fourier变换做出包络阶次谱,从中提取出滚动轴承故障特征阶次。试验结果验证了该方法对于滚动轴承故障特征提取的有效性。

Aiming at non-stationary vibration signal and weak early fault feature during modulated velocity processes,the order tracking is combined with energy operator(EO).A fault diagnosis method for rolling bearings is proposed based on order tracking energy operator(OTEO)and singular value decomposition(SVD).Firstly,the denoising is carried out by SVD for domain vibration signal to reduce noise interference.Secondly,the denoised signal is demodulated and analyzed by OTEO,namely,the signal is demodulated by EO and then order tracking is used to process envelope signal.Finally,FFT is adopted to obtain envelope order spectrum,so that the fault feature order of rolling bearings is extracted.The experimental results verify that the validity of proposed method for fault feature extraction of rolling bearings.