Faults Diagnosis of Rolling-Element Bearings Based on Fourier Decomposition Method and Teager Energy Operator

Abstract

Bearing failure is a common cause of machinery breakdown, resulting in lost production and accidents. Therefore, the early and automatic diagnosis of bearing faults is attracting increasing amounts of attention, in particular through vibration analysis. The vibration signals caused by rotating machines are multi-component and modulated signals; it is essential to demodulate after the components decomposition. Through the previous challenges, we have proposed an approach combining the most effective decomposition and demodulation techniques, namely the Fourier decomposition method (FDM) and the Teager energy operator (TEO). Next, we verified the FDM–TEO method’s efficacy by applying it to synthetic and real bearing vibration signals and comparing it to other well-known methods using the FCFR assessment indicator. Comparative results were quite close, with the FDM–TEO and the empirical mode decomposition both having marginally better mean FCFRs than the variational mode decomposition method.