Sep 2026· International Journal of Acoustics and Vibration· Vol 31, pp. 299· 0 citations
TL;DR
Results show that this method demonstrates superior noise immunity and the ability to diagnose faults without differentiating between operating conditions, and improving the model's ability to represent features effectively.
Abstract
To reduce the impact of complex operating conditions and noise interference on bearing fault diagnosis, this paper proposes a Multi-scale One-dimensional Convolutional Neural Network Model based on First-layer Wide Convolution, CBAM attention and ACON-C activation function (MWCACNN). This approach features a large-kernel, multi-scale feature extractor (WKInception) designed specifically for one-dimensional signals. It connects multiple large-kernel convolutional layers with pooling layers in parallel to extract fault features across different scales, thereby maximizing the retention of global and detailed information from the original signal. Compared to a wide convolutional CNN (WDCNN) model in the first layer, this method emphasizes capturing fine details early and combining them with global features, as confirmed by ablation studies. Additionally, the Convolutional Block Attention Module (CBAM) is used to dynamically enhance the importance of weights of key fault features and their details. The ACON-C activation function is integrated to modify feature nonlinearity adaptively, improving the model's ability to represent features effectively. Diagnostic experiments were conducted to validate this method's effectiveness under complex operating conditions and high noise interference, with further verification using the HIT aviation engine bearing dataset. Results show that this method demonstrates superior noise immunity and the ability to diagnose faults without differentiating between operating conditions.
A bearing fault diagnosis method based on the Dual Attention Multi-Scale Convolutional Neural Network that achieves a peak diagnostic accuracy of 99.69% in cross-condition tasks and maintains accuracy above 92% under −5 dB SNR, outperforming BiTCN-Transformer, MTFC-T, and CTCA in accuracy, recall, and F1-score, while a...
Experimental results demonstrate that the proposed MSODNet model effectively captures both the divergence and commonality of signals under multi-condition scenarios, achieving an average accuracy of 99.35% in bearing vibration fault diagnosis, confirming the superior diagnostic performance and robustness of the MSODNet...
Ke-Ming Liu, Song-Yang Han· International Conference on...· 0 citations
To enhance the reliability and accuracy of bearing fault diagnosis under challenging operating conditions, this study proposes a method that integrates a combined fast Fourier transform (FFT) and variational mode decomposition (VMD) preprocessing framework with a convolutional neural network (CNN)-bidirectional long sh...
Jie Li, Yu-Ying Wang, Jie Yang et al.· Engineering Research Express· 0 citations
To address the challenges of low diagnostic accuracy, limited feature extraction capability and poor cross-condition generalisation in rolling bearing fault diagnosis under small-sample scenarios, this study proposes an end-to-end CWT-MCNN-SNN framework that integrates the continuous
wavelet transform (CWT), a multi-s...
Fei-Long Zhang, Jun Gao, Xing Shao et al.· Insight - Non-Destructive Te...· 0 citations
To address the limitations in model generalization caused by discrepancies in the distribution of multi-source data and insufficient labeling in the fault diagnosis of high-speed train bearings, this paper constructs an intelligent diagnostic computing pipeline designed for cross-operating-condition scenarios. This iss...
Cheng-Xuan Liu, Qi-Rui Liu, Qing Cai· The International Conference...· 0 citations