Abstract
This study presents a new method based on convolutional neural network (CNN) for the gearbox fault identification and classification, which does not need the complex feature extraction process as those traditional recognition algorithms do, and it also depress the uncertainty of arbitrary feature selection. The vibration signals of the gearbox under normal and hybrid fault conditions were collected, and all kinds of signals were transformed to time-frequency images by using S-transform. Then the time-frequency matrices were input to the CNN to classify different types of faults. To evaluate the performance of the CNN, other two deep learning algorithms, deep belief network (DBN) and stacked auto-encoder (SAE), were adopted to classify the gearbox faults for comparison. Experiment results demonstrated that CNN can be effectively used for fault classification.

This publication has 10 references indexed in Scilit: