TorbeamNN: Machine learning based steering of ECH mirrors on KSTAR
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908841732472832 |
|---|---|
| author | Rothstein, Andrew Kim, Minseok Woo, Minho Cha, Minsoo Byun, Cheolsik Kim, Sangkyeun Erickson, Keith Lee, Youngho Josephy-Zack, Josh Butt, Jalal Shousha, Ricardo Joung, Mi Juhn, June-Woo Lee, Kyu-Dong Kolemen, Egemen |
| author_facet | Rothstein, Andrew Kim, Minseok Woo, Minho Cha, Minsoo Byun, Cheolsik Kim, Sangkyeun Erickson, Keith Lee, Youngho Josephy-Zack, Josh Butt, Jalal Shousha, Ricardo Joung, Mi Juhn, June-Woo Lee, Kyu-Dong Kolemen, Egemen |
| contents | We have developed TorbeamNN: a machine learning surrogate model for the TORBEAM ray tracing code to predict electron cyclotron heating and current drive locations in tokamak plasmas. TorbeamNN provides more than a 100 times speed-up compared to the highly optimized and simplified real-time implementation of TORBEAM without any reduction in accuracy compared to the offline, full fidelity TORBEAM code. The model was trained using KSTAR electron cyclotron heating (ECH) mirror geometries and works for both O-mode and X-mode absorption. The TorbeamNN predictions have been validated both offline and real-time in experiment. TorbeamNN has been utilized to track an ECH absorption vertical position target in dynamic KSTAR plasmas as well as under varying toroidal mirror angles and with a minimal average tracking error of 0.5cm. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_11648 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TorbeamNN: Machine learning based steering of ECH mirrors on KSTAR Rothstein, Andrew Kim, Minseok Woo, Minho Cha, Minsoo Byun, Cheolsik Kim, Sangkyeun Erickson, Keith Lee, Youngho Josephy-Zack, Josh Butt, Jalal Shousha, Ricardo Joung, Mi Juhn, June-Woo Lee, Kyu-Dong Kolemen, Egemen Plasma Physics We have developed TorbeamNN: a machine learning surrogate model for the TORBEAM ray tracing code to predict electron cyclotron heating and current drive locations in tokamak plasmas. TorbeamNN provides more than a 100 times speed-up compared to the highly optimized and simplified real-time implementation of TORBEAM without any reduction in accuracy compared to the offline, full fidelity TORBEAM code. The model was trained using KSTAR electron cyclotron heating (ECH) mirror geometries and works for both O-mode and X-mode absorption. The TorbeamNN predictions have been validated both offline and real-time in experiment. TorbeamNN has been utilized to track an ECH absorption vertical position target in dynamic KSTAR plasmas as well as under varying toroidal mirror angles and with a minimal average tracking error of 0.5cm. |
| title | TorbeamNN: Machine learning based steering of ECH mirrors on KSTAR |
| topic | Plasma Physics |
| url | https://arxiv.org/abs/2504.11648 |