1. The document proposes a deep convolutional neural network (DCNN) based data fusion method to identify the health state of planetary gearboxes using raw vibration data from horizontal and vertical sensors. 
2. The method aims to extract features automatically from the fused data with less dependence on expert experience or complex signal processing compared to other methods.
3. Experimental results on data from a planetary gearbox test rig show the proposed DCNN fusion method achieves better identification results than models using individual data streams or other shallow learning methods.

A Deep Convolutional Neural Network Based Fusion Method of Two-Direction Vibration Signal Data For Health State Identification of Planetarygearboxes PDF

The list of books you might like