Rolling Bearing Fault Diagnostic Method Based on EEMD and CHMM Signal Processing

According to the rationale of resonance demodulation of vibration signals of rolling bearing faults, the paper probes into aperiodic vibration signals of rolling bearing faults with variable speed, and the changing rotation rate-related resonance frequency of rolling bearing. An experimental scheme is designed for vibration signals of rolling bearing with variable speed. A rolling bearing fault model is created where frequency conversion is the modulation signal and the frequency conversion resonance attenuation signal is the carrier. A variable speed rolling bearing fault diagnostic method is proposed basing on EHMM (as the fault feature extraction method) and CHMM (as the fault mode recognition method). The feasibility of the method is verified by experiments.