Pitting Detection in a Worm Gearbox Using Artificial Neural Networks

Diagnosis of worm gears faults using vibration analysis is difficult, for this reason; there have been quite little publications, although worm gears are used significant machines in assorted industrial fields. Whenever a defect occurs in a worm system (e.g. pitting, abrasive wear) the performances of the gears deteriorate. Therefore, transmission of motion and power cannot be transferred as demanded. As a result, occurrence of fatal defects becomes inevitable. This paper focuses upon the early detection of localized pitting damages in a worm gearbox using artificial neural networks (ANN) and vibration analysis. Worm gear vibrations are acquired from an experimental rig utilizing a 1/15 worm gearbox. Statistical parameters of vibration signals in the time and frequency domains are used as an input to classifier ANN for multi-class recognition.