A Novel Method for Bearing Safety Detection in Urban Rail Transit Based on Deep Learning

The double tapered roller bearing is widely used in urban rail transit, due to its complex structure, the traditional safety detection is difficult to recognize the early weak fault. In order to solve this problem, a deep learning method for safety detection of roller bearing is put forward. In the experiment, vibration signals of bearing are firstly separated into a series of intrinsic mode functions by empirical mode decomposition, then we extracted the transient energy to construct the eigenvectors. In the pattern recognition, deep learning method is used to generate the safety detector by unsupervised study. There are three states of rolling bearings in experiments, as normal, inner fault and outer fault. The results show that the proposed method is more stable and accurately to identify bearing faults, and the classification accuracy is above 98%.