Unknown fault feature extraction of rolling bearings under variable speed conditions based on statistical complexity measures

As the main transmission components of rotating machinery, rolling bearings have important research significance for fault diagnosis and state detection. However, the operating environment of mechanical equipment is complex, and the fault characteristic information of rolling bearings is often unknown. Through the complex transmission path, the bearing seat vibration sensor picks up the fault vibration response is weaker, and it is often submerged by strong noise. When the speed and load of rotating mechanical equipment change, the fault characteristic of rolling bearings is more obvious under variable speed conditions. The fault diagnosis problem of rolling bearings needs to be solved urgently under strong noise background, variable speed, and unknown fault characteristics. Therefore, this paper studies an unknown fault feature extraction method of variable speed rolling bearing based on statistical complexity measures (SCM). Order analysis preprocesses the variable speed vibration signal of rolling bearings. It is convenient for subsequent fault feature extraction and analysis. The SCM selects the optimal intrinsic mode function (IMF) component corresponding to the Empirical mode decomposition (EMD) decomposition, and it is also evaluated index for the optimal response of stochastic resonance. Therefore, the adaptive frequency shift stochastic resonance effectively extracts the unknown fault features of rolling bearings under strong noise background.