Abstract
In allusion to the problems for the low signal-noise ratio of the monitoring parameters, difficulty of fault characterizating, lack of acceleration life test mode of the space ball bearing and so on, firstly, an acceleration test plan based on temperature and rotational speed compositive step is designed, and a test system which can realize state monitoring, achieves the data set of vibration and temperature during the test is set up, then based on the ensemble empirical mode decomposition (EEMD) method, the original vibration data and noise data are separately decomposed into intrinsic mode function (IMF) components with different scales. with the help of correlation coefficient algorithm, the original vibration data is denoised and the signal-to-noise ratio is raised. Secondly, the principal component analysis (PCA) is used to carry out the fault feature extraction and dimension reduction analysis of the vibration data that is denoised, and then the fault feature data which is characterized by the eigenvectors is got. Finally, the trend analysis of the fault feature data is performed by the support vector machine (SVM) algorithm to obtain the fault feature degradation law based on temperature and rotational speed, and thus it can provide a basis for the evaluation of acceleration life test.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have