Vibration-Acoustic Signal Analysis Of Rolling Bearing

Rolling bearing is widely used in rotating machinery and plays a key role in the machine. Rollingbearing failure may result in huge economic loss or an accident. So it is essential to research onbearing condition monitoring and fault diagnos is. Now vibration analysis is mainly applied in bearingcondition monitoring and fault diagnosis. But in the case of high temperature and high corrosion, it isdifficult to install vibration sensors. Therefore vibration analysis is limited. Sound signal is not subjectto the limitation. It is non-contact measurement, easy to detect early failure. And it does not need toinstall sensors. Therefore it is meaningful to carry out fault diagnosis based on sound signal.Comprehensively using vibration and acoustic analysis technology can improve the reliability of faultdiagnosis, and it is significant for fault diagnosis.The details of this paper are listed as follows:1. The relationship between vibration and noise of rolling bearing, mechanism of vibration andnoise are presented. The vibration-acoustic experiment of the beam verifies the relationship betweenvibration and noise. Essentially, sound signal is effective as well as vibration signal, which isvulnerably interfered with the background and its SNR is low.2. The vibration-acoustic signal of rolling bearing is non-stationary signal. The traditionalenvelope analysis has weaknesses that it need to pre-determine filter center frequency and band widthartificially. So this paper combines EMD with envelope analysis to diagnose bearing and uses EMD toachieve the purpose of filtering, because EMD is an adaptive analysis method, which is suitable forthe non-stationary signal.3. The rolling bearing vibration-acoustic signal analysis software is developed by using GUIinterface tools in Matlab, which is fast and convenient. The effectiveness of the software has beenconfirmed by the experimental data of bearing.4. During the vibration-acoustic experiment of rolling bearing, vibration signal and sound signalare collected at the same time. Analyze those signals with the software above. The result shows thatanalyzing sound signal can diagnose faults of bearing effectively. Comprehensive analysis ofvibration-acoustic signal can improve the reliability of fault diagnosis.