Fault Identification of Bearing Inner Ring Based on Local Feature Scale Decomposition of Segmented Structure
In order to improve the feature extraction ability of bearing vibration signals,a segmented structure local feature scale decomposition(PPLCD)method is designed.According to the performance of kurtosis and correlation coefficient,a kurtosis correlation coefficient(K-C)weight evaluation index is constructed.CWRU bearing parameters are selected as test objects,and a test platform is built.The results show that the effective components obtained by PPLCD decomposition reach smaller MAE and RMSE,and at the same time form a larger correlation coefficient,which is the same as that of the outer ring fault analysis.The PPLCD decomposition method can obtain better performance than the initial LCD.The research can be extended to other fields of mechanical transmission and has great popularization value.
NETL
NSTL
万方数据
