Contrastive adversarial domain adaptation for machine remaining useful life prediction
Enabling precise forecasting of the remaining useful life (RUL) for machines can reduce maintenance cost, increase availability, and prevent catastrophic consequences. Data-driven RUL prediction methods have already achieved acclaimed performance. However, they usually assume that the training and t...
| Main Authors: | , , , , , , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/157026 |
| _version_ | 1811679115253645312 |
|---|---|
| author | Mohamed Ragab Chen, Zhenghua Wu, Min Foo, Chuan Sheng Kwoh, Chee Keong Yan, Ruqiang Li, Xiaoli |
| author2 | School of Computer Science and Engineering |
| author_facet | School of Computer Science and Engineering Mohamed Ragab Chen, Zhenghua Wu, Min Foo, Chuan Sheng Kwoh, Chee Keong Yan, Ruqiang Li, Xiaoli |
| author_sort | Mohamed Ragab |
| collection | NTU |
| description | Enabling precise forecasting of the remaining useful life (RUL) for machines can reduce maintenance cost, increase availability, and prevent catastrophic consequences. Data-driven RUL prediction methods have already achieved acclaimed performance. However, they usually assume that the training and testing data are collected from the same condition (same distribution or domain), which is generally not valid in real industry. Conventional approaches to address domain shift problems attempt to derive domain-invariant features, but fail to consider target-specific information, leading to limited performance. To tackle this issue, in this article, we propose a contrastive adversarial domain adaptation (CADA) method for cross-domain RUL prediction. The proposed CADA approach is built upon an adversarial domain adaptation architecture with a contrastive loss, such that it is able to take target-specific information into consideration when learning domain-invariant features. To validate the superiority of the proposed approach, comprehensive experiments have been conducted to predict the RULs of aeroengines across 12 cross-domain scenarios. The experimental results show that the proposed method significantly outperforms state-of-the-arts with over 21% and 38% improvements in terms of two different evaluation metrics. |
| first_indexed | 2024-10-01T03:04:01Z |
| format | Journal Article |
| id | ntu-10356/157026 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T03:04:01Z |
| publishDate | 2022 |
| record_format | dspace |
| spelling | ntu-10356/1570262022-05-30T06:12:54Z Contrastive adversarial domain adaptation for machine remaining useful life prediction Mohamed Ragab Chen, Zhenghua Wu, Min Foo, Chuan Sheng Kwoh, Chee Keong Yan, Ruqiang Li, Xiaoli School of Computer Science and Engineering Engineering::Computer science and engineering Domain Adaptation Deep Learning Enabling precise forecasting of the remaining useful life (RUL) for machines can reduce maintenance cost, increase availability, and prevent catastrophic consequences. Data-driven RUL prediction methods have already achieved acclaimed performance. However, they usually assume that the training and testing data are collected from the same condition (same distribution or domain), which is generally not valid in real industry. Conventional approaches to address domain shift problems attempt to derive domain-invariant features, but fail to consider target-specific information, leading to limited performance. To tackle this issue, in this article, we propose a contrastive adversarial domain adaptation (CADA) method for cross-domain RUL prediction. The proposed CADA approach is built upon an adversarial domain adaptation architecture with a contrastive loss, such that it is able to take target-specific information into consideration when learning domain-invariant features. To validate the superiority of the proposed approach, comprehensive experiments have been conducted to predict the RULs of aeroengines across 12 cross-domain scenarios. The experimental results show that the proposed method significantly outperforms state-of-the-arts with over 21% and 38% improvements in terms of two different evaluation metrics. Agency for Science, Technology and Research (A*STAR) Submitted/Accepted version This work was supported in part by the A*STAR Industrial Internet of Things Research Program under the RIE2020 IAF-PP Grant A1788a0023, and in part by the National Natural Science Foundation of China under Grant 51835009. The work of Mohamed Ragab was supported by A*STAR SINGA Scholarship. Paper no. TII-20-1888. 2022-04-30T07:13:21Z 2022-04-30T07:13:21Z 2020 Journal Article Mohamed Ragab, Chen, Z., Wu, M., Foo, C. S., Kwoh, C. K., Yan, R. & Li, X. (2020). Contrastive adversarial domain adaptation for machine remaining useful life prediction. IEEE Transactions On Industrial Informatics, 17(8), 5239-5249. https://dx.doi.org/10.1109/TII.2020.3032690 1551-3203 https://hdl.handle.net/10356/157026 10.1109/TII.2020.3032690 2-s2.0-85097333789 8 17 5239 5249 en A1788a0023 IEEE Transactions on Industrial Informatics 10.21979/N9/FMUP9M © 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. The published version is available at: https://doi.org/10.1109/TII.2020.3032690. application/pdf |
| spellingShingle | Engineering::Computer science and engineering Domain Adaptation Deep Learning Mohamed Ragab Chen, Zhenghua Wu, Min Foo, Chuan Sheng Kwoh, Chee Keong Yan, Ruqiang Li, Xiaoli Contrastive adversarial domain adaptation for machine remaining useful life prediction |
| title | Contrastive adversarial domain adaptation for machine remaining useful life prediction |
| title_full | Contrastive adversarial domain adaptation for machine remaining useful life prediction |
| title_fullStr | Contrastive adversarial domain adaptation for machine remaining useful life prediction |
| title_full_unstemmed | Contrastive adversarial domain adaptation for machine remaining useful life prediction |
| title_short | Contrastive adversarial domain adaptation for machine remaining useful life prediction |
| title_sort | contrastive adversarial domain adaptation for machine remaining useful life prediction |
| topic | Engineering::Computer science and engineering Domain Adaptation Deep Learning |
| url | https://hdl.handle.net/10356/157026 |
| work_keys_str_mv | AT mohamedragab contrastiveadversarialdomainadaptationformachineremainingusefullifeprediction AT chenzhenghua contrastiveadversarialdomainadaptationformachineremainingusefullifeprediction AT wumin contrastiveadversarialdomainadaptationformachineremainingusefullifeprediction AT foochuansheng contrastiveadversarialdomainadaptationformachineremainingusefullifeprediction AT kwohcheekeong contrastiveadversarialdomainadaptationformachineremainingusefullifeprediction AT yanruqiang contrastiveadversarialdomainadaptationformachineremainingusefullifeprediction AT lixiaoli contrastiveadversarialdomainadaptationformachineremainingusefullifeprediction |