In recent years, Academia and industry have conducted extensive and in-depth research on bearing-fault-diagnosis technology. However, the current modeling of time–space coupling characteristics in rolling bearing fault diagnosis remains inadequate, and the integration of multi-modal correlations requires further improvement. To address these challenges, this paper proposes a joint diagnosis framework integrating graph convolutional networks (GCNs) with attention-enhanced bidirectional gated recurrent units (BiGRUs). The proposed framework first constructs an improved K-nearest neighbor-based spatio-temporal graph to enhance multidimensional spatial–temporal feature modeling through GCN-based spatial feature extraction. Subsequently, we design an end-to-end spatio-temporal joint learning architecture by implementing a global attention-enhanced BiGRU temporal modeling module. This architecture achieves the deep fusion of spatio-temporal features through the graph-structural transformation of vibration signals and a feature cascading strategy, thereby improving overall model performance. The experiment demonstrated a classification accuracy of 97.08% on three public datasets including CWRU, verifying that this method decouples bearing signals through dynamic spatial topological modeling, effectively combines multi-scale spatiotemporal features for representation, and accurately captures the impact characteristics of bearing faults.
Under no-load conditions, the wet clutch of vehicles generates drag torque across a wide speed range, which increases power loss in the transmission system and significantly impacts its efficiency and reliability. To address the clutch drag issue over a wide speed range, this study first establishes a low-speed drag torque model that simultaneously considers the viscous friction effects in both the complete oil film region and the oil film rupture zone of the friction pair clearance. Subsequently, by solving the fluid-structure interaction dynamics model of the friction plates, the collision force between high-speed friction pairs and the resulting friction torque are determined, forming a method for calculating high-speed collision-induced drag torque. Building on this, a unified drag torque model for wet clutches across a wide speed range is developed, integrating both viscous and collision-induced drag torques. The validity of the wide-speed-range drag torque model is verified through experiments. The results indicate that as oil temperature and friction pair clearance increase, the drag torque decreases and the rotational speed corresponding to the peak drag torque is reduced, while the onset of collision phenomena occurs earlier. Conversely, with an increase in oil supply flow rate, the drag torque rises and the rotational speed corresponding to the peak drag torque increases, but the onset of collision phenomena is delayed. Finally, with the optimization objectives of minimizing the peak drag torque in the low-speed range and the total drag torque at the maximum speed in the high-speed range, an optimization design model for the surface grooves of the clutch friction plates is constructed. An optimized groove pattern is obtained, and its effectiveness in suppressing drag torque across a wide speed range is experimentally validated.
Time series forecasting serves a critical function in domains such as energy, meteorology, and power systems by leveraging historical data to predict future trends. However, existing methods often prioritize long-term dependencies while neglecting the integration of local features and global patterns, resulting in limited accuracy for short-term predictions of non-stationary multivariate sequences. To address these challenges, this paper proposes a time series forecasting model named VBTCKN based on variational mode decomposition and a dual-channel cross-attention network. First, the model employs variational mode decomposition (VMD) to decompose the time series into multiple frequency-complementary modal components, thereby reducing sequence volatility. Subsequently, the BiLSTM channel extracts temporal dependencies between sequences, while the transformer channel captures dynamic correlations between local features and global patterns. The cross-attention mechanism dynamically fuses features from both channels, enhancing complementary information integration. Finally, prediction results are generated through Kolmogorov–Arnold networks (KAN). Experiments conducted on four public datasets demonstrated that VBTCKN outperformed other state-of-the-art methods in both accuracy and robustness. Compared with BiLSTM, VBTCKN reduced RMSE by 63.32%, 68.31%, 57.98%, and 90.76%, respectively.