Empirical mode decomposition of acoustic signals for diagnosis of faults in gears and rolling element bearings

Abstract

Rolling element bearings and gears are the most important components of rotating machines. One of the major causes of machine down time is because of the failure of these elements. Down time of rotating machines can be reduced by monitoring vibration and acoustic behaviour of machine elements. This study describes the application of the empirical mode decomposition (EMD) method to diagnose the faults in rolling element bearings and helical gears. By using EMD, a complicated signal can be decomposed into a number of intrinsic mode functions (IMFs) based on the local characteristic timescale of the signal. The IMFs reveal the intrinsic oscillation modes embedded in the signal. Acoustic signals acquired from the bearings and gears have been decomposed and kurtosis values are extracted from these IMFs to quantify various faults. Results demonstrate the advantages of EMD method to detect the faults in the early stage.