Fault Diagnosis of Rolling Bearing Based on DBN Network
In order to improve the accuracy of rolling bearing fault diagnosis,a rolling bearing fault diagno-sis method based on DBN network is proposed.Aiming at the technical difficulties such as shallow neural networks that are difficult to extract deep fault features from vibration signals and are prone to dimensional disasters,a rolling bearing fault diagnosis model based on DBN network is established by combining the characteristics of deep belief network(DBN)that can process high-dimensional nonlinear data and effec-tively extract fault features.Through verification analysis,the network parameters such as the number of hidden layers,optimal data type,and activation function of DBN were determined,It provides a new method and idea for the setting of DBN network parameters.and the reconstruction ability of the restricted Boltz-mann machine was verified and analyzed.DBN is compared and analyzed with BP,ELM,PN and other shallow neural networks,and the results show that DBN network has high diagnostic accuracy and strong stability,which proves the effectiveness of DBN network in rolling bearing fault diagnosis.
deep belief networkrestricted boltzmann machinesrolling bearingfault diagnosis
NETL
NSTL
万方数据
维普
