The VMD-scale space based hoyergram and its application in rolling bearing fault diagnosis

Abstract

The fast kurtogram is one of the most commonly applied methods for detecting rolling element bearing faults, and has been proven to be to be effective in most cases. However, the shortcomings of the kurtosis index limit the robustness and universality of the method, in that the kurtosis index is sensitive to single impulses with large amplitudes, and the segmentation of the spectrum is not adaptive to different signals. Moreover, fast kurtogram segments the spectrum evenly, which makes the method less adaptive. Therefore, a new diagnosis method for denoising signals, known as the VMD-Scale Space Based hoyergram, is proposed in this paper. Firstly, parameter-optimized Variation Mode Decomposition (VMD) is applied to the signal to calculate the center frequency of each sub-signal, referred to as the Intrinsic Mode Function (IMF). Secondly, the spectrum of the vibration signal is smoothed by means of scale space theory, and the local minimum between each two center frequencies is determined as the boundary. Thirdly, the specific filter center frequency and filter bandwidth are obtained via hoyergram, in which the kurtosis index of the fast kurtogram is replaced by the Hoyer index, and the spectrum is segmented based on its distribution. The frequency band with the largest Hoyer index value contains the impact information. Finally, the periodicity impact can be observed in the corresponding envelope spectrum. The proposed method improves on the traditional fast kurtogram. The analysis results for both simulated and experimental signals verify the effectiveness of the proposed method in rolling bearing fault diagnosis.