Research On Rolling Bearing Fault Diagnosis Method Based On Parameter Optimization VMD

Rolling bearing is a key component of rotating machinery.Its operation state directly affects the stability and safety of rotating machinery.If there is a fault during operation,if it is not found and intervened in time,it will affect the normal operation of rotating machinery at least,and lead to major equipment failure,huge economic loss or endanger personal safety at most.Therefore,timely diagnosis of bearing fault has both theoretical significance and practical value.Therefore,the fault diagnosis of rolling bearing is carried out.The main research contents are as follows:(1)This dissertation uses this as a starting point to study the failure mechanism of rolling bearings and conduct in-depth research on fault signal processing methods and pattern recognition methods,and explore the appropriate signal processing and pattern recognition methods to achieve the purpose of accurate fault diagnosis.This thesis explains some main failure forms of rolling bearing,and expounds how to classify the collected signals and sample the signals.(2)This paper introduces the background of the subject,the technical methods of fault diagnosis at home and abroad and the development status of modal decomposition in detail,analyzes the basic mechanism,failure form,characteristic frequency representation and the established fault impact model of rolling bearing,deeply studies the methods such as EMD and VMD and the phenomenon of modal aliasing,points out that the ability of VMD to suppress modal aliasing is better than the traditional algorithm,and summarizes its existing problems,It provides a theoretical basis for the follow-up fault diagnosis.(3)Aiming at the problem that the early fault characteristics of rolling bearing vibration signal are weak and the impact components representing the fault characteristics are often submerged in noise,resulting in the low accuracy of rolling bearing fault identification,a rolling bearing fault diagnosis model based on MCKD and VMD-AR and LS-SVM is proposed.Firstly,the necessity of combining VMD with MCKD and AR is analyzed,and a fault feature extraction method based on MCKD and VMD-AR model is proposed.Firstly,MCKD is used to reduce the early fault signal of rolling bearing in order to highlight the signal-to-noise ratio of the signal.Then the AR model is established for the reconstructed signal of the sensitive IMF component decomposed by VMD.Finally,the correlation dimension of AR model autoregressive parameter sequence is calculated to extract the early fault features of rolling bearing,calculate the feature vector,input it into PSO-SVM classifier,and finally test the results.The comparative experimental simulation and data verification show that the accuracy of rolling bearing fault diagnosis is greatly improved by this method.Finally,the correlation dimension of AR model autoregressive parameter sequence is calculated to extract the early fault features of rolling bearing,calculate the feature vector,input it into PSO-SVM classifier,and finally test the results.The comparative experimental simulation and data verification show that the accuracy of rolling bearing fault diagnosis is greatly improved by this method.Through theoretical discussion,simulation analysis and laboratory test,the fault diagnosis of rolling bearing is completed.The results show that the method proposed in this thesis can effectively improve the accuracy and expand the fault diagnosis method of rolling bearing.