Open-Set Fault Diagnosis for Industrial Rotating Machines Based on Trustworthy Deep Learning
Authors:
- 11 February 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Cyber-Physical Systems
- Vol. 3, 181-189
Abstract
Detecting and diagnosing faults in rotating machines is crucial for ensuring the safety and reliability of modern industrial cyber-physical systems. Traditional data-driven fault diagnosis methods have achieved significant success when dealing with a set list of known faults and working conditions. However, they become inaccurate and overconfident when faced with new fault classes outside the training set. This paper introduces a novel Evidential Abstention Classifier based on trustworthy deep learning. It can quantify prediction uncertainty and recognize new fault classes without the need for their training data. Experiment results validated the efficacy of the proposed L1 regularization in improving uncertainty quantification. They also highlighted the proficiency of the designed auxiliary training method in generating fault-discriminative features and establishing effective decision boundaries for new fault types. EAC enables accurate open-set fault diagnosis with reduced reliance on historical data, offering improved transparency in the diagnostic process.Keywords
Funding Information
- Future Energy Systems through Canada First Research Excellence Fund (FES-T11-P01, FES-T14-P02)
- Natural Sciences and Engineering Research Council of Canada (RGPIN-2021-02900)
- China Scholarship Council (201806070147)
This publication has 35 references indexed in Scilit:
- Ensemble deep learning: A reviewEngineering Applications of Artificial Intelligence, 2022
- Towards trustworthy machine fault diagnosis: A probabilistic Bayesian deep learning frameworkReliability Engineering & System Safety, 2022
- Data-driven fault diagnosis analysis and open-set classification of time-series dataControl Engineering Practice, 2022
- Dual adversarial network for cross-domain open set fault diagnosisReliability Engineering & System Safety, 2022
- Real-time accurate detection of wind turbine downtime - An Irish perspectiveRenewable Energy, 2021
- A review of uncertainty quantification in deep learning: Techniques, applications and challengesInformation Fusion, 2021
- Open Set Learning with Counterfactual Images2018
- Amplitudes of characteristic frequencies for fault diagnosis of planetary gearboxJournal of Sound and Vibration, 2018
- A subspace learning-based feature fusion and open-set fault diagnosis approach for machinery componentsAdvanced Engineering Informatics, 2018
- A sparse auto-encoder-based deep neural network approach for induction motor faults classificationMeasurement, 2016