Application of multiple spectrum domain fault probability fusion model in high-speed rail bearings

Abstract High-speed railway bearings are one of the core transmission components of high-speed railway multiple units, and their degraded performance can directly affect the safety performance of vehicles. How to identify faults in high-speed rail bearings and represent them in data has always been an industry challenge. To this end, multiple frequency domain fault probability fusion models have been proposed for research. The first choice for this model is to provide an analysis method for converting the frequency domain spectrum to fault probability. Subsequently, five frequency domain spectrum analysis methods that are easy to detect bearing faults were selected, and all of them were converted into fault probability. Finally, based on D-S evidence theory, fusion was carried out to obtain multiple frequency domain fault probability fusion models, achieving efficient identification and data representation of high-speed railway bearing faults. Finally, the effectiveness of the model was verified through fault simulation experiments.