Rolling bearing fault diagnosis method based on frequency domain window empirical wavelet resonance demodulation

Abstract

The invention discloses an adaptive frequency domain window empirical wavelet transformation resonance demodulation method for rolling bearing fault diagnosis. The method comprises steps that step 1, according to acquired parameter indexes, upper and lower cut-off frequency change fluctuation scopes of a frequency domain window are determined; step 2, an empirical wavelet function is constructed, a wavelet coefficient after empirical wavelet transformation is acquired through calculation, and a modal component signal is reconstructed; step 3, an improved envelope harmonic wave signal to noise ratio of the modal component signal is finally determined through normalization processing; step 4, the improved envelope harmonic wave signal to noise ratio of the modal component signal is taken as an optimal fitness function value, a particle swarm optimization method is employed, and the optimal frequency domain window position is determined; and step 5, the fault characteristic information of a rolling bearing fault vibration signal is extracted, resonance demodulation analysis for rolling bearing fault diagnosis is accomplished. The method is advantaged in that a signal frequency band zone can be flexibly selected, and the rolling bearing fault is diagnosed through resonance demodulation through utilizing particle swarm optimization frequency domain window empirical wavelet transformation.

Description

technical field [0001] The invention relates to the technical field of mechanical fault diagnosis and signal processing, in particular to a rolling bearing fault diagnosis method based on frequency domain window empirical wavelet resonance demodulation. Background technique [0002] Rolling bearings are one of the key components in rotating machinery, and their working conditions determine whether the entire mechanical system can operate reliably. However, in practical engineering, the vibration signal of rolling bearing fault is a typical nonlinear and non-stationary signal, and the fault features in the signal are easily covered by various background noises. Therefore, it is very difficult to diagnose rolling bearing faults under strong background noise. [0003] The resonance demodulation method is one of the basic methods of rolling bearing fault diagnosis. During the rolling bearing rotation, the damage point repeatedly collides with the surface of other components in c...

AI Technical Summary

Problems solved by technology

However, in the prior art, there is no relevant technical record that can well solve these three key problems at the same time.

Images