Research On Fault Diagnosis Technology Of Rolling Bearing Based On Vibration Signal

Rolling bearings are a key component of mechanical equipment,and their working conditions affect the operation of the entire equipment.At the same time,rolling bearings are one of the most vulnerable components.Every year,production accidents caused by rolling bearing failures occur in the world,causing huge property losses.Therefore,it is very necessary to study t he fault diagnosis of rolling bearings.In this paper,a fault diagnosis and data management system is developed for the monitoring of the running condition of rolling bearings.Firstly,this paper describes the working principle,vibration mechanism and characteristic frequency of rolling bearings.Then,the extraction method of rolling bearing fault signal characteristics is studied.Decompose the denoised vibration signal using wavelet packet decomposition algorithm and CEEMDAN algorithm respectively.Taking sub-band energy or mode component energy as an initial fault feature vector.Using the ReliefF algorithm or correlation analysis method,some features that contribute less to the classification are discarded,and the final fault characteristics are obtained.Secondly,the classification method of rolling bearing fault characteristics is studied.Combined with the classification principle of support vector machine method,the genetic algorithm and particle swarm optimization algorithm are used to optimize the penalty factor parameters and kernel function parameters of support vector machine.The support vector machine classifier is trained by using the processed bearing fault feature data to classify and diagnose the unknown state signal.And the different feature extraction algorithms and feature classification algorithms are combined to compare the classification accuracy of different combinations,and finally the optimal combination method is selected.Finally,the fault diagnosis and data management system of rolling bearing was developed by using Matlab and LabView joint programming method.The software includes rolling bearing vibration signal,real-time monitoring and storage of working condition parameters,bearing model management,curve display and timefrequency analysis of bearing vibration data,and intelligent classification diagnosis of bearing faults.According to the needs of the project,a fault diagnosis test bench was set up.A simulation experiment was carried out to simulate its operation at different speeds and different types of damage.The vibration signals in the experiment are collected and analyzed by the software system developed in this paper.The software system developed in this paper is used to diagnose the rolling bearings in operation.The experimental results verify that the software system has good performance and is of great significance for ensuring the safe operation of bearings in mechanical equipment.