Research On Rolling Bearing Performance Degradation Characterization And Remaining Useful Life Prediction Method

Rolling bearing is the key component in the mechanical equipment transmission system.Because it is in the extremely severe operating environment such as heavy load,high speed and high temperature,it is easy to fail,and then cause system level failure.Therefore,it is the key to ensure the safe and reliable operation of mechanical equipment to master the performance degradation state and remaining useful life of rolling bearing.This paper takes the performance degradation characterization and remaining useful life prediction of rolling bearing as the research subject,and carries out the bearing life cycle test.On this basis,based on the vibration signal of the bearing,the research on feature extraction,performance degradation characterization and remaining useful life prediction is carried out.The main contents are as follows:(1)Taking 6207 deep groove ball bearing(material: GCr15)as the test object,the rolling bearing accelerated life test bench was used to conduct a full life cycle performance degradation test to monitor,record and analyze the characteristic parameters of the state monitoring quantity(temperature,vibration)throughout And the change rule provides a reference basis for the selection of subsequent performance degradation behavior characterization parameters.(2)Aiming at the problem of feature extraction of vibration signals,a feature set reflecting bearing degradation is constructed.71 feature parameters related to the signs of bearing performance degradation were extracted from the time domain,frequency domain and time-frequency domain to form the original feature set.The results show that in the life cycle of the bearing,the original features show different types of change trends,representing the distinct information about the degradation process,which can fully and effectively reflect the degradation information of the bearing.(3)Construct a health index to characterize the degradation state of rolling bearing performance.Using correlation,monotonicity and robustness as feature evaluation indexes,the sensitive features reflecting the degradation of bearing performance are selected,and based on the PCA method,multiple sensitive features are fused to construct a health index that characterizes the degradation of bearing performance.It is verified through experiments that the constructed health index can effectively characterize the degradation state of the bearing in three stages of normal operation,slight degradation and severely degradation.(4)Construct an EMD-Kriging model to predict the remaining useful life of rolling bearings.First,the EMD method is used to extract the main trend of the health index;secondly,the remaining useful life of the bearing is predicted based on the Kriging model;finally,the feasibility and effectiveness of the proposed model are verified by comparison and analysis with typical prediction methods where the same dataset is used.The research in this paper provides a method for the characterization of rolling bearing performance degradation and remaining useful life prediction,which has important engineering significance for improving the reliability and security of rolling bearing and even the transmission system of mechanical equipment.