A review of stochastic resonance in rotating machine fault detection

Applications of stochastic resonance to machinery fault detection: A review and tutorial

Multi-scale deep intra-class transfer learning for bearing fault diagnosis

Fault detection of wind turbine based on SCADA data analysis using CNN and LSTM with attention mechanism

Fault diagnosis of rotating machines based on the EMD manifold

Multi-Scale Stochastic Resonance Spectrogram for fault diagnosis of rolling element bearings

It is not easy to identify nascent defect of a bearing by analyzing the vibration data because of the disturbance of background noise. With the help of stochastic resonance (SR) technology, the weak and unrecognizable output signal of a non-stationary mechanical system can be enhanced with the aid of the additional noise of the system. However, the effect of current SR methods still has room for improvement, especially for identifying sensitive fault information in non-stationary signals. This paper presents a new method called multi-scale SR spectrogram (MSSRS) for bearing defect diagnosis. The new method realizes the SR in every scale of the time-frequency distribution of the input signal and so is able to well deal with the non-stationary transient signal. The proposed method can also highlight the defect-induced frequency component by selecting a proper scale corresponding to the impulse information. We analyzed the experimental bearing vibration data with defect information using the new proposed method, the result has shown that there is a better effectiveness with our method than the former SR methods in bearing defect identification.

Enhao Wu

Papers

Multi-Scale Stochastic Resonance Spectrogram for fault diagnosis of rolling element bearings

Multi-scale stochastic resonance spectrogram for rolling element bearing defect diagnosis