Application of Empirical Mode Decomposition for Impulsive Signal Extraction to Detect Bearing Damage – Industrial Case Study

Abstract

Bearings damage detection is one of the most important topic in condition monitoring. Main problem in industrial application of bearing vibration diagnostics is the masking of informative signal by interfering signals. It requires the usage of techniques based on advanced signal enhancement in order to extract useful diagnostic components from the measured vibration signals. The paper shows application of Empirical Mode Decomposition (EMD) in extraction of weak impulsive signal from raw vibration signals generated by complex mechanical systems employed in the industry (driving units of belt conveyors). Impulsive character of the vibration signals is very often associated with a mechanical fault. The purpose of this processing is decomposition of the signal in order to detect impacts related to the damages in rolling element bearings (REB).