A Multiscale Feature Fusion Graph Convolutional Network for Cross-Domain Learning in Bearing Fault Detection

In the industrial scene, mechanical equipment often faces dynamic changes of speed, load, and other working conditions, so accurate diagnosis of rolling bearing faults is still a key challenge in industrial condition monitoring, in which the scarcity of marker data and cross-domain differences seriously limit the generalization of the model. Conventional unsupervised domain adaptation (UDA) approaches predominantly focus on data classification and domain-specific annotation, often neglecting the extraction of inherent structural characteristics present in the data. To address this issue, we propose the implementation of a multiscale feature fusion graph convolutional network (GCN) for UDA. The classifier and the domain discriminator are responsible for processing class-related and domain-specific label information, respectively. Regarding structural representation, a multiscale convolutional architecture is first utilized to derive feature representations from the input signals, producing the feature matrix (X). In the subsequent phase, the fusion graph generation layer (FGGL) investigates the latent structural patterns of samples by generating a topological matrix (A). The sample graph representation is subsequently constructed using the multireference feeler-graph convolutional network (MRF-GCN) architecture. During the final stage, the maximum mean discrepancy (MMD) metric is employed to reduce domain discrepancy through the alignment of structural patterns observed in the sample graphs. The experimental findings demonstrate that the multiscale feature fusion GCN for cross-domain learning (MFGCDL) framework outperforms existing UDA approaches in terms of both stability and precision for rolling bearing fault detection, while successfully capturing transferable cross-domain characteristics.