Vibration and current transient monitoring for gearbox fault detection using multiresolution Fourier transform

This paper deals with an experimental investigation of fault diagnosis in a multistage gearbox under transient loads. An induction motor drives the multistage gearbox, which is connected to a DC generator for loading purpose. The signals studied are the vibration transients, recorded from an accelerometer fitted at the tail-end bearing of the gearbox; and the current transients drawn by the induction motor. Three defective cases and three transient load conditions are investigated. Advanced signal processing techniques such as discrete wavelet transform (DWT) and a corrected multiresolution Fourier transform (MFT) are applied to investigate the vibration and current transients. It is observed from the vibration transients that the load removal is a high-frequency phenomenon. With increase in defect severity, not only the defective gear mesh frequency gains energy, but also large impact energy appears in low-frequency regions. Whereas in the current transients, though load removal is a low-frequency phenomenon, a very small transient is observed at high-frequency regions for defective gears. With increase in defect severity, energy is distributed to the sidebands of the gear mesh frequency across supply line frequency. A statistical feature extraction technique is proposed in order to find a trend in detection of defects. A condition monitoring scheme is devised that can facilitate in monitoring vibration and current transients in the gearbox with simultaneous presence of transient loads and defects. (C) 2007 Elsevier Ltd. All rights reserved.