Research On Fault Diagnosis Method Of Rolling Bearing Based On CEEMD-SVD And ELM

Rolling bearing is one of the core components of rotary machinery equipment.Its running status is directly related to the stability and safety of rotary machinery equipment.Therefore,fault diagnosis of rolling bearings has important theoretical research value and economic significance for the operation and maintenance of rotating machinery and equipment.Feature extraction and fault recognition of fault information is the key to fault diagnosis of rolling bearings.Because the vibration signals of rolling bearings are affected by ambient noise and transmission paths and other factors of the equipment,they exhibit nonlinearity and non-stationarity,which greatly increases the difficulty of fault diagnosis.In view of this,this paper uses Complementary Ensemble Empirical Mode Decomposition(CEEMD),Singular Value Decomposition(SVD),and Extreme Learning Machine(ELM)to extract signals from feature processing and feature extraction.From the perspective of classification and fault diagnosis,an experimental study on the method of fault diagnosis of rolling bearings is conducted.The main research work of this paper is as follows:(1)In order to solve the problem that the fault information of rolling bearing is difficult to be extracted,the fault feature extraction method of rolling bearing based on EEMD and SVD is proposed.Firstly,the vibration signal of the rolling bearing is decomposed by EEMD method,and several Intrinsic Mode Functions(IMF)are decomposed to select the sensitive component by using the kurtosis to reconstruct the signal so as to highlight the local characteristics of the vibration signal.Then,the reconstruction is based on this reconstruction.The signal is constructed into a Hankel matrix,and the matrix is subjected to SVD processing to filter out background noise and extract fault feature information.Experimental results show that the feature extraction method combining EEMD and SVD can effectively extract the fault feature information of rolling bearings.(2)For the problem that EEMD extracts the fault feature information of rollingbearings,there is a problem of residual white noise and low computational efficiency.A fault feature extraction method of rolling bearing combined with CEEMD and SVD is proposed.Firstly,the CEEMD method is used to decompose the bearing vibration signal to obtain several IMF components.The correlation coefficient method is used to select the IMF component(eliminate false components),and the filtered effective component signal is reconstructed;then the IMF component matrix is subjected to SVD processing.,to achieve the fault diagnosis of rolling bearings.The CEEMD method not only can effectively solve the problem that white noise added in the EEMD decomposition process cannot be completely neutralized,but also improves the efficiency of iterative operations.Comparing with EEMD-SVD method,the experimental results show that this method can further improve the efficiency of fault signal feature extraction of rolling bearings.(3)Based on the feature extraction of CEEMD-SVD,an extreme learning machine(ELM)is used as a fault identification method,and a rolling bearing fault diagnosis method combining CEEMD-SVD and ELM is proposed.The method firstly performs CEEMD-SVD analysis on the vibration signal of the rolling bearing and extracts its singular value as the characteristic parameter to construct the feature vector.Then,it uses the constructed eigenvector to complete the training of the ELM fault classification model;finally,it uses the vibration of the rolling bearing in different running states.The test data completes the ELM classification.The experimental results show that the fault recognition rate of this method for rolling bearings reaches 95%,which can effectively achieve the accurate diagnosis of rolling bearing faults.This paper takes the rolling bearing in the rotating machinery as the research object,completes its feature extraction and exploration methods of fault diagnosis methods,enriches the research theory of fault diagnosis of rolling bearings,and promotes the application and development of the rolling bearing fault diagnosis technology.