Haihong Tang

Heavy background noise inevitably masks the weak fault characteristics of low-speed bearings in offshore wind turbines, severely degrading the fault detection accuracy of conventional spectral amplitude modulation. To address this challenging issue, a novel hybrid diagnosis method integrating fast spectral coherence and spectral amplitude modulation is proposed for low-speed bearing fault detection. Firstly, spectral amplitude modulation is implemented on raw vibration signals to construct modified signal components. Subsequently, the fast spectral coherence algorithm is employed to eliminate complex noise interference and reconstruct high-quality enhanced envelope spectra. Further squaring transformation and amplitude normalization operations are conducted to effectively amplify incipient and weak fault signatures buried in noisy signals. Multiple visual analysis strategies are also adopted to intuitively present the fault diagnosis results. The effectiveness and superiority of the proposed method are comprehensively validated by simulated signals with varying signal-to-noise ratios and experimental data of bearing inner and outer race faults under a low rotating speed of 60 RPM. The results demonstrate that the proposed hybrid method significantly remedies the inherent limitations of traditional diagnosis methods, providing a robust and reliable technical solution for incipient fault monitoring and health assessment of low-speed bearings operating in harsh offshore wind field environments.

To address the issue of difficult extraction of bearing fault features caused by weak fault features and strong environmental noise in low-speed, a low-speed bearing fault diagnosis method based on wavelet threshold denoising and spectral amplitude modulation is proposed. The proposed method effectively overcomes the limitation that the traditional spectral amplitude modulation is greatly affected by noise in low-speed. Firstly, the raw signal is subjected to wavelet threshold denoising to reduce the interference of strong background noise, thereby obtaining the denoised signal. Secondly, the denoised signal is subjected to spectral amplitude modulation to enhance the bearing fault impulses. Finally, the envelope spectrum is normalized to facilitate the visual display of fault feature frequencies. The proposed method is analyzed through simulated and experimental signals in low-speed. The experimental results indicate that the proposed method can reduce noise interference and effectively extract fault features in low-speed.

To address the challenge of extracting effective fault features at low speeds, where fault information is weak and heavily influenced by environmental noise, a parameter-adaptive variational mode decomposition (VMD) method is proposed. This method aims to overcome the limitations of traditional VMD, which relies on manually set parameters. The sparrow search algorithm is used to calculate the fitness function based on mean envelope entropy, enabling the adaptive determination of the number of mode decompositions and the penalty factor in VMD. Afterward, the optimised parameters are used to enhance traditional VMD, enabling the decomposition of the raw signal to obtain intrinsic mode function components. The kurtosis criterion is then used to select relevant intrinsic mode functions for signal reconstruction. Finally, envelope analysis is applied to the reconstructed signal, and the results reveal the relationship between fault characteristic frequencies and their harmonics. The experimental results demonstrate that compared with other advanced methods, the proposed approach effectively reduces noise interference and extracts fault features for diagnosing low-speed bearing faults.

The present research on intelligent bearing fault diagnosis assumes that the same feature distribution is used to obtain training and testing data. However, the domain shift (distribution discrepancy) issue generally occurs in both datasets because of different operational conditions. The domain adaptation techniques are preferably applied for fault diagnosis to handle the domain shift issue. Moreover, collecting sufficient testing data or labelled data in real industries is a challenging task. Therefore, the multi-kernel joint distribution adaptation (MKJDA) with dynamic distribution alignment is proposed for bearing fault diagnosis. This method dynamically joins both the marginal and conditional distributions and uses the multi-kernel to solve the non-linear problems to extract the most effective and robust representation for cross-domain issues. Moreover, it runs with the unlabelled task domain to perform the diagnosis by iteratively updating the pseudo code. The experimental results (two public datasets and one experimental dataset) demonstrated that the proposed method (MKJDA) exhibited stable and robust accuracy while conducting bearing fault diagnosis. It can effectively address the most crucial issue: intelligent diagnosis methods must re-train the model when the distribution differs between the source domain (the model is learned) and the target domain (the learned model is applied).

Adaptive intelligent fault diagnosis technology can improve the automatic status monitoring capability and the success rate of fault diagnosis. This article proposed an adaptive mode decomposition and noise reduction algorithm (VAMD) that use a variable spectral segmentation framework to optimize analytical mode decomposition (AMD) to automatically decompose the mode information in rotating machinery signals. The framework relies on the variability of the window width and envelope estimation characteristics of the order statistics filter (OSF) to increase the diversity of the center frequencies and bandwidth. A novel harmonic correlation index (HCI) is designed to identify the characteristics of rotating machinery faults from various levels of results and improve the usability in mechanical equipment fault diagnosis. The proposed method has successfully achieved fault diagnosis of rolling bearing.