Multisensor Feature Fusion Based Rolling Bearing Fault Diagnosis Method

To fully utilize the fault information and improve diagnosis accuracy of rolling bearings, a multisensor feature fusion method is proposed. The contains two steps. First, intrinsic mode function (IMF) each sensor vibration signal calculated by variational decomposition (VMD), redundant such as noise eliminated. Then, time-domain, frequency-domain multiscale entropy features are extracted based on preferred IMF fused into one multidomain dataset. In second step, deep autoencoder network (DAEN) constructed first step used input DAEN, further classified. experimental results show that proposed model has higher classification compared with existing methods.