A Deep Ensemble Learning Model for Rolling Bearing Fault Diagnosis

Abstract

Rolling bearing is an important component of rotating machinery. The accurate fault diagnosis of rolling bearing is very important. Nowadays, experts began to explore the combination strategies of deep learning networks. Ensemble learning can achieve higher recognition accuracy by combining multiple models. Therefore, a deep ensemble learning model is proposed for rolling bearing fault diagnosis. Firstly, four different Convolutional Neural Networks (CNN) networks are constructed as the base-learners. Secondly, the 4-fold cross validation method is adopted for training the base-learner. Finally, the Artificial Neural Network (ANN) is used as the meta-learner and the stacking method is used for model ensemble. The proposed method can get high classification accuracy and accurately identify all kinds of faults.