Research on Bearing Fault Diagnosis Based on ResNet-BLSTM-SVM

With the rapid development of our society, rail transit such as metro has become the main means of transportation. As the core component of rail vehicles, the health status of bearings directly influences the smooth operation of the rail vehicles. A Resnet18-BLSTM-SVM model is proposed to effectively extract the fault signal features of rolling bearings and accurately identify bearing faults. In order to verify the superiority of this model, a rolling bearing acoustic emission test bench is established. Acoustic emission signals from four different state bearings are collected and divided into 3200 samples, split into training and test sets in the ratio of 8:2. Overall, the results show that the proposed model can achieve higher accuracy in a smaller number of epoches than the other three models.