Multi-component fault diagnosis of wheelset-bearing using shift-invariant impulsive dictionary matching pursuit and sparrow search algorithm

Abstract Wheelset-bearing is an important part of the bogie for high-speed train. Fault diagnosis of wheelset-bearing is of great significance for the safety of the railway service. In the diagnosis of multi-component fault for wheelset-bearing system, the extraction and separation of each fault component are crucial. Therefore, a novel hierarchical shift-invariant impulsive dictionary matching pursuit method based on sparrow search algorithm is proposed in this paper. This method can extract each fault component by adopting the thought of hierarchy and the shift-invariant structure of dictionary. In order to achieve hierarchically adaptive extraction, sparrow search algorithm is introduced to optimize parameters, and a tuning rule for the optimal sparsity used in matching pursuit is established. Additionally, to validate the superiority of the proposed method, both simulated and experimental signals are analyzed and compared. The results show that the proposed method has good performance in multi-component fault diagnosis of wheelset-bearing.