Weak fault feature extraction using adaptive chirp mode decomposition with sparsity index regrouping scheme and time-delayed feedback stochastic resonance

The failure features of rolling bearings are often weak due to the influence of strong background noise. In addition, the vibration signals of faulty rolling bearing often show nonlinear and non-stationary characteristics, and the conventional time-frequency method is no longer suitable for extracting effective fault features. In order to extract the early weak fault characteristics of rolling bearing accurately, a weak fault feature extraction method for rolling bearing by combining adaptive chirp mode decomposition (ACMD) based on sparsity index regrouping scheme with time-delayed feedback stochastic resonance (TDSR) is proposed in the paper. The proposed method comprehensively utilizes the adaptive decomposition characteristics of ACMD for multi-component non-stationary signals and the enhancement effect of TDFSR on low-frequency signals in the fast Fourier transform (FFT) result. Firstly, ACMD is used to decompose the early weak fault signal of rolling bearing into a series of mode signals, then the proposed signal regrouping scheme based on sparsity index is utilized to regroup the obtained series of modes. Secondly, the optimal reconstructed component containing the main fault information is input into the calculation model of TDSR. Finally, FFT is performed on the output signal of TDSR to extract the fault characteristics effectively.