Research on Bearing Vibration Signal Processing Based on MCMC-SPERMAN Algorith

This paper analyzes the vibration information dataset of a roller bearing under the whole working life cycle., collects the fault information and forms the corresponding wear model., wear modeling of bearing inner rings based on dynamic experimental datasets of vibration is derived by the fusion of Markov Chain Monte Carlo (hereafter referred to as MCMC) algorithm and Spearman correlation coefficient. In the face of the huge signal data problem in the data processing process., it is proposed to reduce the intermediate redundancy and overlapping process in the huge data by changing the sampling data through distributization and introducing the Effective Sample Size (ESS). Compared with the traditional algorithm can reduce the data access volume of 93% −95%., greatly reducing the complexity of the operation., improve the efficiency of the operation., and ensure the accuracy of the model., the resulting data model compared with other data processing algorithms have a better advantage of the actual error accuracy.