Bearing Faulty Prediction Method Based on Federated Transfer Learning and Knowledge Distillation
In this paper, a novel bearing faulty prediction method based on federated transfer learning and knowledge distillation is proposed with three stages: (1) a “signal to image” conversion method based on the continuous wavelet transform is used as the data pre-processing method to satisfy the input ch...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Machines |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-1702/10/5/376 |
| _version_ | 1797498421719859200 |
|---|---|
| author | Yiqing Zhou Jian Wang Zeru Wang |
| author_facet | Yiqing Zhou Jian Wang Zeru Wang |
| author_sort | Yiqing Zhou |
| collection | DOAJ |
| description | In this paper, a novel bearing faulty prediction method based on federated transfer learning and knowledge distillation is proposed with three stages: (1) a “signal to image” conversion method based on the continuous wavelet transform is used as the data pre-processing method to satisfy the input characteristic of the proposed faulty prediction model; (2) a novel multi-source based federated transfer learning method is introduced to acquire knowledge from multiple different but related areas, enhancing the generalization ability of the proposed model; and (3) a novel multi-teacher-based knowledge distillation is introduced as the knowledge transference way to transfer multi-source knowledge with dynamic importance weighting, releasing the target data requirement and the target model parameter size, which makes it possible for the edge-computing based deployment. The effectiveness of the proposed bearing faulty prediction approach is evaluated on two case studies of two public datasets offered by the Case Western Reserve University and the Paderborn University, respectively. The evaluation result shows that the proposed approach outperforms other state-of-the-art faulty prediction approaches in terms of higher accuracy and lower parameter size with limited labeled target data. |
| first_indexed | 2024-03-10T03:33:09Z |
| format | Article |
| id | doaj.art-c0e8347630f34eaaad068e19faa145b2 |
| institution | Directory Open Access Journal |
| issn | 2075-1702 |
| language | English |
| last_indexed | 2024-03-10T03:33:09Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Machines |
| spelling | doaj.art-c0e8347630f34eaaad068e19faa145b22023-11-23T11:53:05ZengMDPI AGMachines2075-17022022-05-0110537610.3390/machines10050376Bearing Faulty Prediction Method Based on Federated Transfer Learning and Knowledge DistillationYiqing Zhou0Jian Wang1Zeru Wang2Computer Integrated Manufacturing System (CIMS) Research Center, College of Electronics and Information Engineering, Tongji University, Shanghai 201804, ChinaComputer Integrated Manufacturing System (CIMS) Research Center, College of Electronics and Information Engineering, Tongji University, Shanghai 201804, ChinaComputer Aided Design (CAD) Research Center, College of Electronics and Information Engineering, Tongji University, Shanghai 201804, ChinaIn this paper, a novel bearing faulty prediction method based on federated transfer learning and knowledge distillation is proposed with three stages: (1) a “signal to image” conversion method based on the continuous wavelet transform is used as the data pre-processing method to satisfy the input characteristic of the proposed faulty prediction model; (2) a novel multi-source based federated transfer learning method is introduced to acquire knowledge from multiple different but related areas, enhancing the generalization ability of the proposed model; and (3) a novel multi-teacher-based knowledge distillation is introduced as the knowledge transference way to transfer multi-source knowledge with dynamic importance weighting, releasing the target data requirement and the target model parameter size, which makes it possible for the edge-computing based deployment. The effectiveness of the proposed bearing faulty prediction approach is evaluated on two case studies of two public datasets offered by the Case Western Reserve University and the Paderborn University, respectively. The evaluation result shows that the proposed approach outperforms other state-of-the-art faulty prediction approaches in terms of higher accuracy and lower parameter size with limited labeled target data.https://www.mdpi.com/2075-1702/10/5/376knowledge distillationfederated transfer learningparameter sizeknowledge transferenceedge-computing deployment |
| spellingShingle | Yiqing Zhou Jian Wang Zeru Wang Bearing Faulty Prediction Method Based on Federated Transfer Learning and Knowledge Distillation Machines knowledge distillation federated transfer learning parameter size knowledge transference edge-computing deployment |
| title | Bearing Faulty Prediction Method Based on Federated Transfer Learning and Knowledge Distillation |
| title_full | Bearing Faulty Prediction Method Based on Federated Transfer Learning and Knowledge Distillation |
| title_fullStr | Bearing Faulty Prediction Method Based on Federated Transfer Learning and Knowledge Distillation |
| title_full_unstemmed | Bearing Faulty Prediction Method Based on Federated Transfer Learning and Knowledge Distillation |
| title_short | Bearing Faulty Prediction Method Based on Federated Transfer Learning and Knowledge Distillation |
| title_sort | bearing faulty prediction method based on federated transfer learning and knowledge distillation |
| topic | knowledge distillation federated transfer learning parameter size knowledge transference edge-computing deployment |
| url | https://www.mdpi.com/2075-1702/10/5/376 |
| work_keys_str_mv | AT yiqingzhou bearingfaultypredictionmethodbasedonfederatedtransferlearningandknowledgedistillation AT jianwang bearingfaultypredictionmethodbasedonfederatedtransferlearningandknowledgedistillation AT zeruwang bearingfaultypredictionmethodbasedonfederatedtransferlearningandknowledgedistillation |