Bearing fault detection based on instantaneous energy spectrum

Abstract

A novel method for detection and diagnosis the bearing fault according to instantaneous energy spectrum (IES) based on empirical mode decomposition (EMD) method is presented. EMD can adaptively decompose the vibration signal into a series of zero mean oscillatory functions called intrinsic mode functions (IMFs). Hilbert transform can track the instantaneous amplitude and instantaneous frequency of the intrinsic mode functions at any instant. Then the instantaneous energy spectrum can be calculated based on the intrinsic mode functions. The instantaneous energy is thus associated with the deterioration in bearing condition and can be utilized for bearing fault detection. The experimental results show that instantaneous energy spectrum is a sensitive indicator of the existence of damage in the gearbox.