Feature Adaptive Fusion Network with Inter-domain Hybrid Weighting Mechanism: A Domain Adaption Method for Cross-domain Bearing Fault Diagnosis

Domain Adaptation (DA) has achieved significant advancements in cross-domain bearing fault diagnosis. However, the variations in working conditions and environmental factors lead to non-discriminative features. These non-discriminative features not only contaminate the realistic representation of discriminative features but also cause negative transfer during the alignment of features between source and target domains. To address the problem, this paper proposes a method termed Feature Adaptive Fusion Network with Inter-domain Hybrid Weighting Mechanism (FAFNIHW). Firstly, a feature adaptive fusion network employs parallel convolutional branches to capture multi-scale contextual features, followed by a dynamic weighting mechanism that adaptively fuses the parallel features, thereby enhancing the diversity of critical discriminative features. Besides, an inter-domain hybrid weighting module is introduced to refine the alignment of features by considering the importance of features across both the channel and spatial dimensions, ultimately, this process prioritizes critical discriminative features while assigning lower priority to non-discriminative ones. Comprehensive experiments on cross-machine fault diagnosis for gearbox bearings between the trolley and hoisting mechanisms of a quayside container crane demonstrate the effectiveness and superiority of FAFNIHW compared to state-of-the-art methods.