Research On Fault Diagnosis Method Of Main Bearing Of TBM Based On Wavelet Packet And Hilbert Envelope Analysis

The main bearing is one of the core components of the shield machine.Its service life determines the performance of the shield machine and restricts the construction schedule.According to the statistics shows that during the use of the main bearing of the shield machine,only 5% can reach its design service life.If the main bearing is damaged in the construction process,it will cause the shield machine to stop.Due to the long replacement cycle and difficult maintenance,the entire project progresses seriously lagging behind and brings huge economic losses.Therefore,real-time fault diagnosis and timely detection of the fault have important significance for the good/healthy operation of the shield main bearing.This paper first introduces the mechanical structure of the main bearing of the shield machine,the common vibration mechanism of the main bearing,the form of the fault,the natural frequency of the bearing,the frequency of the fault feature and the form of the vibration of the bearing.The time domain eigenvalue method and precision of the vibration signal in the fault diagnosis are analyzed.Diagnostic methods,precise diagnostic methods include: wavelet transform method,wavelet packet transform method,Hilbert transform and other methods.In view of the shortcomings of these diagnostic methods,a vibration signal detection method combining wavelet packet and Hilbert envelope analysis is proposed.Secondly,the main bearing vibration fault signal detection scheme and hardware system are designed.The hardware system has the characteristics of serial communication,fast sampling rate and 12-bit AD conversion.And in the Lab VIEW software environment has developed a corresponding supporting software system,the software system with data acquisition,data storage,data communications,report printing,fault diagnosis,time-frequency eigenvalues and simple analysis of frequency domain analysis methods.The real-time monitoring of bearing temperature signal and vibration signal can be realized by analyzing the bearing vibration data obtained by the system and the relevant data of the main bearing temperature and speed of the shield machine collected by SIEMENS 300 PLC.Finally,the fault diagnosis system developed above is used to analyze the data collected by the laboratory bearing fault simulation platform and the field measured data of a certain type of shield bearing made in China.The analysis of the experimental platform data can accurately extract the bearing fault characteristic frequency and determine the bearing fault type.Compare the data collected on site with historical data to find out that the monitored shield machine is working properly.Although no characteristic frequency was detected during bearing failure,a largeamount of available data was still accumulated to provide a basis for assessing the health status of the shield machine and future detection plans,and At the same time,it further verifies the feasibility of this method in practical engineering applications.