Fault prediction model for rolling bearings based on double adaptive sliding time windows

Abstract

To address the problems of traditional and neural network-based methods in rolling bearing fault prediction, we proposed a fault prediction model for rolling bearings based on double adaptive sliding time windows. Firstly, the rolling bearing vibration signal is mapped into fault features that can characterise its degradation state by setting up a state estimation non-linear operator that can remove correlations. Secondly, a loss function is used as a criterion to set up an adaptive update mechanism for the model parameters, as well as a sliding time window capable of adaptively selecting the data length. Finally, the validity of the proposed failure prediction model is verified by simulating the occurrence of failures under the combined sudden and gradual failures in practice using the whole life cycle data of rolling bearings released by Xi'an Jiaotong University.