Health Indicator of Bearing Constructed by rms-CUMSUM and GRRMD-CUMSUM With Multifeatures of Envelope Spectrum

Because a fault of bearing may cause a serious accident of the running machine, it is very important for monitoring condition and identifying the performance degradation of bearing. Two novel health monitoring indicators are proposed by root mean square with cumulative sum (rms-CUMSUM) and growth rate of real-time Mahalanobis distance with cumulative sum (GRRMD-CUMSUM) based on multifeatures of envelope spectrum for health condition monitoring of bearings. Multifeature parameters of envelope spectrum are integrated with rms and RMD based on the vibrational signal of bearings, and the GRRMD is calculated. The rms and GRRMD are processed by the CUMSUM method. The rms-CUMSUM and GRRMD-CUMSUM can represent the tendency and rate of bearings' performance degradation, respectively. The different stages of bearings' performance degradation can be classified using the initial fault time and the special turning point of GRRMD-CUMSUM reference to rms-CUMSUM. The initial fault time is roughly estimated by the abrupt warning point of RMD, and further accurately determined by the backtracking strategy with the interval-halving principle by fast independent component analysis based on empirical mode decomposition (EMD-Fast ICA). The proposed approach was verified by two different experimental data sets. The experimental results show that the proposed approach is simple and efficient for determining the initial fault, assessing, and classifying the performance degradation of bearings.