基于包络解调随机共振和CEEMD的机械早期微弱故障诊断方法研究

针对噪声背景下机械振动信号早期故障特征提取难题,提出一种基于包络解调随机共振和互补总体经验模态分解的机械早期微弱故障提取及诊断新方法。首先对含噪声机械故障信号进行包络解调处理,然后对包络信号进行变尺度随机共振输出,使故障特征信号得到增强,最后对处理后的信号进行互补总体经验模态分解(CEEMD),得到机械振动信号故障特征分量,实现故障特征提取及诊断。对机械故障诊断实例表明,该方法不仅能增强信号幅值,同时减少了虚假分量,提高了CEEMD算法的精度,有效提取出被噪声淹没的微弱故障信号,提高了机械早期故障诊断效果。

Study on Mechanical Early Weak Fault Diagnosis Method Based on CEEMD and Envelope Demodulation Stochastic Resonance

Aiming at the early fault feature extraction problem of mechanical vibration signal under noise background, a novel method based on envelope demodulation stochastic resonance and CEEMD is proposed . With the method,the mechanical fault signal with noise is processed by envelope demodulation,and then through stochastic resonance system the rescaling signals are enhanced. Finally the output result is decomposed by CEEMD,obtaining the fault feature components to realize feature extraction and fault diagnosis. The rolling bearing fault diagnosis example shows that the method can not only improve the signal amplitude and reduce the false component,but also improve CEEMD algorithm precision and effectively extract fault signal submerged in noise.