Research On Fault Diagnosis And State Recognition Of Rolling Bearing Based On Vibration Signal

Rolling bearing is one of the most widely used and most easily damaged components in rotating machinery,Its operating state has a significant impact on the equipment itself and production activities,In order to ensure the smooth running of the equipments and reduce the occurrence of the accidents,it is of great significance to deeply developing the research on the fault diagnosis and the state detection technology for rolling bearing.In this paper,the rolling bearing is taken as the research object,starting from the analysis of vibration signal,the time-frequency signal processing method of variational mode decomposition and the pattern recognition method of support vector machine are combined to study a fault diagnosis and state recognition method for rolling bearing.Firstly,the fault mechanism and common analysis methods of rolling bearings are described in detail in this paper.The structure and types of rolling bearings are simply explained with related graphs.On the basis of analyzing the vibration mechanism of rolling bearings,the fault types and causes,fault characteristic frequencies and common analysis methods of rolling bearings are analyzed in detail.Finally,the characteristics of vibration signals under different fault states are introduced with actual bearing signals.Secondly,the time-frequency signal processing method of variational mode decomposition(VMD)is studied and applied to the early fault diagnosis of rolling bearings.On the basis of explaining the relevant principle of VMD,in order to show the effect of this method on signal processing,the simulation signal is constructed for the problem of signal mode aliasing and the empirical mode decomposition(EMD)method is introduced to compare with it.By comparing the results of the two methods,the superiority of the variational mode decomposition method is shown.Aiming at the parameter determination of VMD in use,a method combining observing center frequency and local kurtosis maximum optimization is proposed,and the method is verified in the early fault diagnosis of rolling bearings.Thirdly,the parameter optimization method of support vector machine(SVM)based on multi-population genetic algorithm(MPGA)is studied,and the MPGA-SVM model for rolling bearing state classification is created.For the parameters of support vector machine model selection problem considering using genetic algorithm to optimize,to compare the standard genetic algorithm and multi-population genetic algorithm optimization effect,through the optimization examples confirmed multi-population genetic algorithm optimization results better,the multi-population genetic algorithm was applied to the parameters of the SVM model optimization problem,which is established for rolling bearing state classification of MPGA-the SVM model.Finally,a rolling bearing fault diagnosis method based on VMD-SVD and MPGASVM is proposed.In view of the excelent performance of the variational mode decomposition method in fault feature extraction,a fault feature extraction method based on VMD-SVD is introduced by combining the singular value decomposition method with it,combining the fault feature extraction method with MPGA-SVM constructed in the previous section,a rolling bearing fault diagnosis method based on VMD-SVD and MPGA-SVM is proposed.In order to show the application effect of the proposed method,it is applied to the fault diagnosis of two bearing databases.The good fault diagnosis effect proves the effectiveness of the proposed method.In order to facilitate the display and application of this method,the human-computer interaction interface is designed on the basis of the method proposed in this paper.