Search

Aiming at the problem that traditional fast spectral correlation (Fast-SC) cannot effectively extract the fault feature of rolling bearings under strong background noise, we propose to select a band with abundant fault information on the spectral frequency axis by Spectral Gini Index (SGI) and integrate over them to obtain the Improved Enhanced Envelope Spectrum (IEES). The proposed method doesn’t rely on the selection of precise fault characteristic frequency, and has a good industrial application prospect. The simulated and experimental results of rolling bearings prove the effectiveness of the algorithm.

In this paper, relationships between conditional variance statistic and the center frequency as well as the bandwidth of the filter are researched, and we find that conditional variance statistic is suitable for selecting the center frequency of the filter, but undesirable for selecting the bandwidth. Meanwhile, considering that the traditional iterative way to select the optimal resonance frequency band will be affected by the step size, we choose to fix the bandwidth at five times fault characteristic frequency, then Whale Optimization Algorithm (WOA) is utilized to select the optimal center frequency so as to give consideration to both computing efficiency and accuracy of the selection of center frequency. To further suppress the in-band noise, the filtered signal is analyzed by high order energy operator, and the optimal two energy operators are chosen based on Fault Characteristic Index (FCI) for cross-correlation spectrum analysis. Simulation and experimental results indicate that the proposed algorithm can extract the fault feature of rolling bearings under strong background noise effectively.

Vibration signals of rolling element bearings during operation are always very complex, random strongly and broadband. Adaptive Local Iterative Filtering Decomposition (ALIFD) can overcome the smoothness and adaptive flaws of Iterative Filtering Decomposition (IFD), but it is so susceptible to random noise that it’s less effective. Here, spatial correlation was proposed. Firstly, the signal was denoised by spatial correlation and decomposed into several modes by ALIFD. Finally, the envelope demodulation was analyzed to extract fault feature. The simulating signal analysis and bearing fault simulator show that this method can be available for separating different frequencies of bearing fault vibration signals.

There are always the nonlinear and non-stationary characteristics and periodic pulse in vibration signals of rolling element bearings when there are partial faults in those bearings. Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) overcomes the presence of spurious modes and residual noise in Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), but it cannot clearly and accurately extract the weak fault feature of rolling element bearings under the strong background noise. Here, Multipoint Optimal Minimum Entropy Deconvolution Adjusted (MOMEDA) was proposed. A bearing simulator was used to collect vibration signals of bearing inner and outer race, which was enhanced by MOMEDA, decomposed into several Intrinsic Mode Functions(IMFs) by ICEEMDAN, and analyzed by the envelope demodulation, finally gaining the frequency of shaft speed, BPFI (ball pass frequency, inner race) and harmonics, sidebands spaced, BPFO (ball pass frequency, outer race) and harmonics. The results show that this method can be used to accurately extract different frequency components of bearing fault vibration signals and diagnose bearing different fault location.

In order to extract impulse components from bearing vibration signals with strong background noise, a fault feature extraction method based on multi-scale average combination difference morphological filter and Frequency-Weighted Energy Operator is proposed in this paper. The average combination difference morphological filter (ACDIF) is used to enhance the positive and negative impulse components in the signal. The double-dot structure element (SE) is used instead of zero amplitude flat SE to improve the effectiveness of fault feature extraction. The weight coefficients of the filtered results at different scales in multi-scale ACDIF are adaptively determined by an optimization algorithm called hybrid particle swarm optimizer with sine cosine acceleration coefficients (H-PSO-SCAC). At last, as the Frequency-Weighted Energy Operator (FWEO) outperforms the enveloping method in detecting impulse components of signals, the filtered signal is processed by FWEO to extract the fault features of bearings. Results on simulation and experimental bearing vibration signals show that the proposed method can effectively suppress noise and extract the fault features from bearing vibration signals.