Fault Diagnosis of Rolling Bearings Based on Explicit Dynamics and Zoom Spectrum

To explore the dynamic evolution of early bearing failure, typical three-dimensional finite element models of 6205 bearing were established. Based on an explicit algorithm and in single-point integral manner, kinetic analysis of typical faults in inner and outer rings was conducted, considering speed, load, contact and friction conditions. The acceleration signals of defected bearings were extracted and then analyzed by Hilbert envelope demodulation and zoom spectral to extract the signal characteristics of the defected bearing. Simulation results show that under error condition, fault characteristics can be embodied from nodes in different locations of the rolling bearing; fault characteristic frequency can be found by acceleration responses with zoom spectrum. Typical faults of the rolling bearing were identified effectively and further study for bearing failure transfer-path characteristics may be feasible.