Gearbox faults identification using vibration signal analysis and artificial intelligence methods

The paper addresses the implementation of feature based artificial neural networks and vibration analysis for the purpose of automated gearbox faults identification. Experimental work has been conducted on a specially designed test rig and the obtained results are validated on a belt conveyor gearbox from a mine strip bucket wheel excavator SRs 1300. Frequency and time domain vibration features are used as inputs to fault classifiers. A complete set of proposed vibration features are used as inputs for self-organized feature maps and based on the results a reduced set of vibration features are used as inputs for supervised artificial neural networks. Two typical gear failures were tested: worn gears and missing teeth. The achieved results show that proposed set of vibration features enables reliable identification of developing faults in power transmission systems with toothed gears.