Rolling Element Bearing Defect Detection Based On Wavelet Transform Of Sound Signal

Acoustic condition monitoring of machinery has received considerable attention in recent years. The main advantage of using sound signal in fault diagnosis is that the acoustic measurements can be performed at a distance from the machine, and sound signal is sensitive to initial faults. Acoustic detection of roller bearings defects is performed by means of wavelet transform and a new statistical moment. The contributions and conclusions are made as follows:The characteristics of sound signal of rolling bearing are discussed. Emphasis is focused on the relationship between faults and vibration-acoustic signal. Acoustic detection is difficult due to the perception that monitoring of airborne sound from a machine is noisy and complex when it is in a normal industrial environment..The quantitative indices that can be extracted from the sound signals are presented. The root mean square level is used to monitor wearing fault. The skewness and kurtosis of sound signal are examined. A new statistical moment which can replace kurtosis is proposed to monitor surface damage fault. The sensitivity of this moment to changes of load and speed is less than kurtosis. Furthermore, the advantage of this moment over the traditional kurtosis value is verified by the simulate and measured signal.It is pointed out that continuous wavelet transform is useful in weak signal detection which help itself to be used in fault diagnosis. The concept of wavelet entropy is introduced and it is considered as a rule for Morlet wavelet parameter optimization. Moreover, the feature of localized defect bearing can be extracted by means of multi-scale wavelet decomposition of sound signal.A method for acoustic detection of bearings is presented. The root mean square level and the statistical moment are used for bearing condition monitoring. Defect detection is performed by means of Morlet wavelet transform. The effectiveness of proposed method was verified by roller bearing test in anechoic room.