Plasma Confinement State Classification via FPP Relevant Microwave Diagnostics
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917018452623360 |
|---|---|
| author | Clark, Randall Glukhov, Vacslav Subbotin, Georgy Nurgaliev, Maxim Kachkin, Aleksandr Austin, Max Orlov, Dmitri M. |
| author_facet | Clark, Randall Glukhov, Vacslav Subbotin, Georgy Nurgaliev, Maxim Kachkin, Aleksandr Austin, Max Orlov, Dmitri M. |
| contents | We present a parsimonious and robust machine learning approach for identifying plasma confinement states in fusion power plants (FPPs) where reliable identification of the low-confinement (L-mode) and high-confinement (H-mode) regimes is critical for safe and efficient operation. Unlike research-oriented devices, FPPs must operate with a severely constrained set of diagnostics. To address this challenge, we demonstrate that a minimalist model, using only electron cyclotron emission (ECE) signals, can deliver accurate and reliable state classification. ECE provides electron temperature profiles without the engineering or survivability issues of in-vessel probes, making it a primary candidate for FPP-relevant diagnostics. Our framework employs ECE as input, extracts features with radial basis functions, and applies a gradient boosting classifier, achieving high accuracy with test accuracy averaging 96\% correct predictions. Robustness analysis and feature importance study confirm the reliability of the approach. These results demonstrate that state-of-the-art performance is attainable from a restricted diagnostic set, paving the way for minimalist yet resilient plasma control architectures for FPPs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_14078 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Plasma Confinement State Classification via FPP Relevant Microwave Diagnostics Clark, Randall Glukhov, Vacslav Subbotin, Georgy Nurgaliev, Maxim Kachkin, Aleksandr Austin, Max Orlov, Dmitri M. Plasma Physics We present a parsimonious and robust machine learning approach for identifying plasma confinement states in fusion power plants (FPPs) where reliable identification of the low-confinement (L-mode) and high-confinement (H-mode) regimes is critical for safe and efficient operation. Unlike research-oriented devices, FPPs must operate with a severely constrained set of diagnostics. To address this challenge, we demonstrate that a minimalist model, using only electron cyclotron emission (ECE) signals, can deliver accurate and reliable state classification. ECE provides electron temperature profiles without the engineering or survivability issues of in-vessel probes, making it a primary candidate for FPP-relevant diagnostics. Our framework employs ECE as input, extracts features with radial basis functions, and applies a gradient boosting classifier, achieving high accuracy with test accuracy averaging 96\% correct predictions. Robustness analysis and feature importance study confirm the reliability of the approach. These results demonstrate that state-of-the-art performance is attainable from a restricted diagnostic set, paving the way for minimalist yet resilient plasma control architectures for FPPs. |
| title | Plasma Confinement State Classification via FPP Relevant Microwave Diagnostics |
| topic | Plasma Physics |
| url | https://arxiv.org/abs/2510.14078 |