基于迁移学习与可解释分析的轴承故障诊断方法及系统
By using improved particle swarm optimization and SHAP algorithms, transfer learning and interpretability analysis of bearing fault diagnosis models were achieved, solving the problems of decreased accuracy and insufficient adaptability of traditional methods in noisy environments, and improving the accuracy and interpretability of bearing fault diagnosis.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANDONG UNIV
- Filing Date
- 2026-01-05
- Publication Date
- 2026-07-17
AI Technical Summary
Traditional bearing fault diagnosis methods are prone to failure under noise and changing operating conditions. Deep learning models have reduced accuracy in real vehicle environments. Existing transfer learning methods lack adaptability and interpretability, making them difficult to apply in safety-critical scenarios.
An improved particle swarm optimization algorithm is used for hyperparameter optimization, and the SHAP algorithm is combined for interpretability analysis of the transfer learning process. The transfer learning process is analyzed through feature distribution alignment and interpretable algorithms, thereby improving the accuracy of fault classification.
It improves the accuracy and interpretability of bearing fault diagnosis, solves the problem of inter-domain generalization failure, enhances the accuracy and stability of cross-domain feature alignment, and meets the application requirements of safety-critical scenarios.
Smart Images
Figure CN121880966B_ABST