Cyclostationarity of Acoustic Emissions (AE) for monitoring bearing defects

Cyclostationarity is a relatively new technique that offers diagnostic advantages for analysis of vibrations from defective bearings. Similarly the Acoustic Emission (AE) technology has emerged as a viable tool for preventive maintenance of rotating machines. This paper presents an experimental study that characterizes the cyclostationary aspect of Acoustic Emission signals recorded from a defective bearing. The cyclic spectral correlation, a tool dedicated to evidence the presence of cyclostationarity, was compared with a traditional technique, the envelope spectrum. This comparison showed that the cyclic spectral correlation was most efficient for small defect identification on outer race defects though the success was not mirrored on inner race defects. An indicator, based on this cyclostationary technique, has also been proposed. It is concluded that its offers better sensitivity to the continuous monitoring of defects compared to the use of traditional temporal indicators (RMS, Kurtosis, Crest Factor).