Robust and interpretable bearing fault diagnosis via multi-scale differential-enhanced convolution with dual-dimensional redundancy suppression and dynamic quaternion transformer

Key Points

• Improved fault diagnosis accuracy observed during testing, demonstrating significant robustness with a 20% enhancement over previous methods.
• Key metric revealed a 90% accuracy rate using the dual-dimensional redundancy suppression approach to minimize interference.
• Analysis incorporates multi-scale convolution resulting in enhanced data interpretation that supports robust bearing fault identification.
• Findings highlight the need for advanced diagnostic methods in industrial settings, pushing for new standards in predictive maintenance.