改进EBCLS模型在滚动轴承故障预测中的应用OA

Aiming at the problem of long training time and low accuracy in predicting rolling bearing faults of high-speed locomotives,an enhanced width convolutional learning system(EBCLS)based on fast iterative fusion of residual learning with enhanced nodes is proposed.The system first extracts signal features by combining the broad learning system(BLS)with convolutional neural networks,and continuously optimizes and updates weights by integrating residual learning and adding enhanced nodes during the training process.Finally,the trained model is used to predict the data with sliding windows and output the prediction results.The proposed method was validated and compared with other BLS methods for prediction results,and the results showed that this method exhibited better real-time prediction performance while improving prediction accuracy.