Xueyi Li

Fault Diagnosis of Gearbox Bearings Under Extreme Class Imbalance Based on Multi-Resolution Windows and Density-Aware Cross-Modal Fusion

Robust bearing fault diagnosis under variable speeds and extreme class imbalance remains challenging due to the difficulty in decoupling transient-periodic features. To address this, we propose a Dual-Branch Weighted Convolution Network with Cross-Attention Fusion (DBWC). Specifically, a dual-branch architecture with spatial density-weighted kernels is proposed to decouple high-frequency transients from low-frequency periodic trends. Subsequently, a Cross-Attention Fusion module synthesizes these heterogeneous features by using global contexts to filter local noise. Additionally, an imbalance-aware strategy integrating Focal Loss and composite augmentation is developed to mitigate model bias. Extensive experiments on MCC5-THU and HUST benchmarks demonstrate that DBWC achieves an accuracy of 91.00% and 90.12%, respectively. The proposed method outperforms state-of-the-art models by an average margin of 5%, providing a data-efficient paradigm for complex industrial monitoring.

MAML-S3M: Selective state space meta-learning for cross-condition few-shot bearing fault diagnosis

Deep learning achieves widespread success in fault diagnosis. However, its effectiveness is hindered in practical industrial environments due to complex operating conditions and data sparsity. This article proposes a novel model-agnostic meta-learning framework based on a selective state space model (MAML-S3M) to address the challenge of cross-condition few-shot bearing fault diagnosis. The framework introduces three core innovations. First, the continuously stacked selective state space module dynamically adjusts its receptive field, enabling precise feature extraction under varying conditions. Second, the channel attention mechanism is combined with the selective state space model to capture multi-scale features, thereby enhancing the feature extraction capability of the model. Third, the introduction of an explicit information discarding strategy during meta-task optimization refines the meta-learning process, thereby yielding optimal parameters. Extensive experiments on bearing datasets across different operating conditions demonstrate that the proposed MAML-S3M achieves superior diagnostic accuracy, with an average accuracy of 99.18% across six cross-condition tasks, outperforming state-of-the-art methods such as generalized model-agnostic meta-learning (GMAML) by at least 1.1%. The improvements are particularly helpful in scenarios with complex operating conditions and scarce samples, maintaining over 94% accuracy even in the challenging “10-way 1-shot” setting. We have made the paper’s results publicly available on GitHub. The link is as follows: https://github.com/12138250/MAML-S3M .

A federated class-incremental learning framework with dynamic client participation and evolution for machine fault diagnosis