Accurate diagnosis of bearing faults is crucial for ensuring the safe and reliable operation of rotating machinery. To enhance the recognition accuracy of rolling bearings under nonlinear and non-stationary vibration conditions, this study proposes an integrated approach combining a multi-stage signal preprocessing strategy, termed EMDWS (Empirical Mode Decomposition with Wavelet denoising and SMOTE), with a hybrid deep learning architecture that integrates a Convolutional Neural Network (CNN) and a Transformer model, hereinafter referred to as CNT (CNN-Transformer). The method first applies empirical mode decomposition (EMD) in conjunction with wavelet denoising to enhance the representation of non-stationary fault features. Subsequently, the synthetic minority oversampling technique (SMOTE) is employed to address the issue of class imbalance in the dataset. A hybrid CNN-Transformer model is constructed by integrating convolutional neural networks and Transformer modules, enabling the extraction of both local and global signal characteristics. Furthermore, Bayesian optimization is applied to fine-tune the model’s hyperparameters, thereby enhancing both the efficiency and robustness of the model. Experimental results demonstrate that the proposed method achieves a high identification accuracy of 99.83%, indicating its effectiveness in distinguishing various bearing fault types.

Bearing fault diagnosis for rotating machinery remains challenging due to strong background noise and variable operating conditions in harsh industrial environments. To address these issues, this paper proposes a robust and lightweight bearing fault diagnosis framework, termed Multi-scale and Dual-scale Wavelet Mamba (MDW-Mamba), for noisy and non-stationary industrial applications. The proposed method employs a multi-stage “noise-feature decoupling” strategy, comprising a statistically guided three-channel input, a Multi-scale Omni-kernel Convolutions (MOM-Conv) backbone, and a Dual-Scale Wavelet Mamba (DSW-Mamba) module. This architecture explicitly separates fault components from broadband noise while enabling efficient long-term dependency modeling. Experimental results on three public datasets under variable-speed conditions demonstrate that MDW-Mamba consistently outperforms state-of-the-art methods. It achieves an average classification accuracy exceeding 87% even at a signal-to-noise ratio of −6 dB, showing exceptional robustness against strong interference. Furthermore, the model maintains an ultra-lightweight footprint of 0.26 M parameters and a fast inference time of 1.285 ms, indicating strong potential for real-time, on-board deployment in resource-constrained industrial edge devices.