Adaptive Weighted Feature Mode Decomposition and Its Application in Bearing Fault Diagnosis

A Rolling Bearing Fault Diagnosis Method Using Adaptive Decomposition and Impact Feature Enhancement Fusion

A novel fault diagnosis method based on adaptive signal decomposition and intelligent classification integration is proposed, outperforming other comparative models and exhibiting superior diagnostic precision and robustness.

Spectral trend-guided empirical wavelet transform and its application in wheelset-bearing fault diagnosis

Wheelset-bearing is a critical component for ensuring the safe and stable operation of high-speed trains, and its fault diagnosis is of great importance. Empirical wavelet transform (EWT) offers an adaptive decomposition framework for wheelset-bearing fault diagnosis, but its performance heavily depends on the selection of spectral segmentation boundaries. To address this limitation, this paper proposes a novel adaptive decomposition method, named spectral trend-guided empirical wavelet transform (STGEWT), which integrates the proposed spectral domain Fourier transform-based spectral trend extraction (SDFTSTE) and the multi-band refined ratio of cyclic content (MRRCC) metric. The SDFTSTE captures the low-frequency content of the spectral signal by removing its high-frequency components, thereby yielding a smooth spectral trend curve that serves as the basis for determining the segmentation boundaries. The MRRCC selects the fault-related components by integrating amplitude information over multiple harmonic bands of fault characteristic frequencies. The effectiveness of the proposed STGEWT is validated using both simulated and experimental vibration signals. Comparative analyses with other methods demonstrate that STGEWT consistently achieves superior performance. The results indicate that STGEWT can be regarded as an effective and reliable tool for wheelset bearing fault diagnosis under complex operating conditions.

Fast multivariate all time-scale decomposition method and its application in gear fault diagnosis

Aiming at the limitations of existing multivariate signal decomposition methods such as fast multivariate empirical mode decomposition (FMEMD) and completely adaptive projection multivariate local characteristic-scale decomposition (CAPMLCD) for gear fault diagnosis, this paper proposes a fast multivariate all-time-scale decomposition (FMATD) method. FMATD incorporates the ATD as its one-dimensional kernel within an efficient “projection-decomposition-reconstruction” framework, preserving the mode separation capability and adaptivity of ATD. Meanwhile, the efficient decomposition framework enhances computational efficiency and avoids over-decomposition. Furthermore, a fast projection strategy is designed to select the projection vectors in real time based on the signal energy distribution, thereby enhancing computational efficiency and decomposition accuracy. Applying FMATD to gear simulation signals and real vibration signals from faulty face gears demonstrates that the proposed method can effectively extract fault modes from face gear signals. Compared with FMEMD, CAPMLCD, and multivariate variational mode decomposition, FMATD yields the component with the clearest fault features in the envelope spectrum. In terms of computational efficiency, FMATD outperforms both FMEMD and CAPMLCD.

A Novel Narrowband Filtering Demodulation Method Based on Adaptive Multi-Level Spectra Segmentation Strategy and Its Application in Bearing Fault Diagnosis

Rolling bearings, as a key component of rotating machinery, require precise fault diagnosis to ensure the safe and reliable operation of industrial systems. Nevertheless, the performance of traditional narrowband filtering demodulation (NFD) methods is constrained by inherent limitations of spectral segmentation frameworks and insufficient discriminative capability of feature indicators (FIs). To address these limitations, this paper proposes a new NFD method based on an adaptive multi-level spectra segmentation strategy. Firstly, using power spectral density (PSD) as the analysis basis, an iterative framework is constructed to obtain multi-level spectral trend lines (STLs), which achieves multi-perspective characterization of spectral features. Secondly, the local minimum points of the STLs are used as the segmentation boundaries to extract the demodulation frequency band. Subsequently, a robust blind feature indicator, synergistic characterization criterion (SCC), is proposed, which can simultaneously fully evaluate periodicity and impulsiveness, guiding the selection of the optimal demodulation frequency band (ODFB). Finally, based on the enhanced demodulation spectrum, power exponent transformation is introduced to construct a generalized spectral family, and the adaptive determination of the optimal transformation parameter is guided by frequency-domain signal-to-noise ratio (FDSNR), thereby obtaining the generalized enhanced demodulation spectrum (GEDS). Validation experiments on laboratory and public datasets demonstrate that the proposed method outperforms Fast Kurtogram, Autogram, and CFFsgram, with average improvements of 63.86% and 89.06% in mean-peak ratio (MPR) and fault feature coefficient (FFC), respectively, and provides a new perspective for NFD and expands its application potential in bearing fault diagnosis and condition monitoring.

Adaptive smoothed soft-threshold method for noise-robust bearing fault diagnosis

A robust multi-class bearing fault diagnosis framework using envelope analysis, cepstrum prewhitening and machine learning

The proposed framework provides an effective balance between diagnostic accuracy, robustness, interpretability, and computational efficiency, making it a promising solution for intelligent condition monitoring and predictive maintenance of rotating machinery.