Deep-Learning and Dynamic Time Warping-Based Approaches for the Diagnosis of Reactor Systems
Hoejun Jeong, Jihyun Kim, Doyun Jung, Jangwoo Kwon, Jiawei Xiang · 2024
Read Deep-Learning and Dynamic Time Warping-Based Approaches for the Diagnosis of Reactor Systems by Hoejun Jeong, Jihyun Kim, Doyun Jung, Jangwoo Kwon, Jiawei Xiang (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11644858.txt.
Description
Hoejun Jeong and colleagues combine deep learning with dynamic time warping to diagnose a hidden degradation inside nuclear reactors. The core support barrel, an internal structure, is held by clamping force whose degradation can significantly affect plant safety and reliability. Earlier studies could detect that degradation occurred but not pinpoint its size and position.
The method applies time warping to the magnitude of ex-core neutron noise signals in the frequency domain, letting the model learn how sensor values change with damage state. Autoencoder-based representation learning extracts features while preventing overfitting, improving robustness. Experiments show the size and position of clamping force degradation can be accurately predicted. Internal structures cannot be inspected directly while a plant runs. The approach turns vibrations a reactor already emits into a map of its own wear.
Catalog record
| CREATOR | Hoejun Jeong, Jihyun Kim, Doyun Jung, Jangwoo Kwon, Jiawei Xiang |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |