Yuxuan Ban

Accurately extracting faulty sound signals from belt conveyor rollers within the high-noise environment of coal mine operations presents a formidable challenge. To address this issue, this study introduces an innovative fault diagnosis method that merges the variational modal de-composition (VMD) model with the Swin Transformer deep learning network model. First, the study employed the adaptive VMD method to eliminate intense noise from the original signal of the rollers, while also assessing the reconstruction accuracy of the VMD signal across different modal components. Subsequently, we delved into the impact of the parameter structure of the Swin Transformer network model on the fault diagnosis accuracy. Finally, the accuracy of the method was validated using a sound test dataset from the rollers. The results indicated that optimizing the K-value of the VMD method effectively reduced the noise in the reconstructed signal, and the Swin Transformer excelled in extracting both local and global features. Specifically, on the conveyor roller sound dataset, it was shown that, after the VMD reconstruction of the signal so that the highest Pearson correlation coefficient corresponded to a modal component of 3 and adjusting the parameters of the Swin Transformer coding layer, the combination of the VMD+Swin-S model achieved an accuracy of 99.36%, while the VMD+Swin-T model achieved an accuracy of 98.6%. Meanwhile, the accuracy of the VMD+Swin-S model was higher than that of the VMD + CNN model combination, with 95.4% accuracy, and the VMD+ViT model, with 97.68% accuracy. In the example application experiments, compared with other models the VMD+Swin-S model achieved the highest accuracy rate at all three speeds, with 98.67%, 98.32%, and 97.65%, respectively. Overall, this approach demonstrated high accuracy and robustness, rendering it an optimal choice for diagnosing conveyor belt roller faults within environments characterized by strong noise.

The operating environment of rolling bearings in mine hoists is complicated, and detecting their faults is hindered by a weak and unstable initial vibration signal. This directly affects the ability to extract pertinent fault features. This paper puts forward an adaptive fault diagnosis method for rolling bearings that combines the Variational Modal Decomposition (VMD) model and Vision Transformer (ViT) deep learning network model. The objective was to address the difficulty of extracting relevant fault features from bearing vibration signals in environments with strong noise levels. First, an improved VMD+ViT model was used to remove the strong noise from the original bearing signal and adaptively classify the fault types; then, the impacts of modal components and encoder numbers on the accuracy of fault diagnosis were explored. Finally, the proposed methodology was validated by applying it to actual rolling bearing fault data, including both open-source and fault test datasets. The research findings indicated that employing a VMD+ViT integrated model consisting of one modal component with the highest Pearson correlation coefficient and eight encoders resulted in high accuracy in diagnosing faults, even in the presence of high levels of noise in the bearing’s vibration signal. The proposed diagnostic method achieved a diagnostic accuracy of over 92.70% on the open-source bearing dataset with strong interference noise and over 98.62% on the fault test dataset. The proposed method exhibited high accuracy and strong robustness, making it suitable for effectively diagnosing and accurately identifying different categories of rolling bearing faults in mine hoists, even in environments with high levels of noise.