Robust confinement state classification with uncertainty quantification through ensembled data-driven methods

Maximizing fusion performance in tokamaks relies on high energy confinement, often achieved through distinct operating regimes. The automated labeling of these confinement states is crucial to enable large-scale analyses or for real-time control applications. While this task becomes difficult to aut...

Full description

Saved in:
Bibliographic Details
Main Authors: Yoeri Poels, Cristina Venturini, Alessandro Pau, Olivier Sauter, Vlado Menkovski, the TCV Team, the WPTE Team
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:Nuclear Fusion
Subjects:
Online Access:https://doi.org/10.1088/1741-4326/adf349
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849404683880235008
author Yoeri Poels
Cristina Venturini
Alessandro Pau
Olivier Sauter
Vlado Menkovski
the TCV Team
the WPTE Team
author_facet Yoeri Poels
Cristina Venturini
Alessandro Pau
Olivier Sauter
Vlado Menkovski
the TCV Team
the WPTE Team
author_sort Yoeri Poels
collection DOAJ
description Maximizing fusion performance in tokamaks relies on high energy confinement, often achieved through distinct operating regimes. The automated labeling of these confinement states is crucial to enable large-scale analyses or for real-time control applications. While this task becomes difficult to automate near state transitions or in marginal scenarios, much success has been achieved with data-driven models. However, these methods generally provide predictions as point estimates, and cannot adequately deal with missing and/or broken input signals. To enable wide-range applicability, we develop methods for confinement state classification with uncertainty quantification and model robustness . We focus on off-line analysis for TCV discharges, distinguishing L-mode, H-mode, and an in-between dithering phase (D). We propose ensembling data-driven methods on two axes: model formulations and feature sets. The former considers a dynamic formulation based on a recurrent Fourier neural operator-architecture and a static formulation based on gradient-boosted decision trees. These models are trained using multiple feature groupings categorized by diagnostic system or physical quantity. A dataset of 302 TCV discharges is fully labeled, and we release it publicly to encourage the community to build upon this work. We evaluate our method quantitatively using Cohen’s kappa coefficient for predictive performance and the expected calibration error for the uncertainty calibration. Furthermore, we discuss performance using a variety of common and alternative scenarios, the performance of individual components, out-of-distribution performance, cases of broken or missing signals, and evaluate conditionally-averaged behavior around different state transitions. Overall, the proposed method can distinguish L, D and H-mode with high performance, can cope with missing or broken signals, and provides meaningful uncertainty estimates.
format Article
id doaj-art-2a1cf210952546e49b7908fb1d8aba23
institution Kabale University
issn 0029-5515
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series Nuclear Fusion
spelling doaj-art-2a1cf210952546e49b7908fb1d8aba232025-08-20T03:36:54ZengIOP PublishingNuclear Fusion0029-55152025-01-0165909602210.1088/1741-4326/adf349Robust confinement state classification with uncertainty quantification through ensembled data-driven methodsYoeri Poels0https://orcid.org/0000-0002-4071-4855Cristina Venturini1https://orcid.org/0009-0005-9873-1171Alessandro Pau2https://orcid.org/0000-0002-7122-3346Olivier Sauter3https://orcid.org/0000-0002-0099-6675Vlado Menkovski4https://orcid.org/0000-0001-5262-0605the TCV Teamthe WPTE TeamÉcole Polytechnique Fédérale de Lausanne (EPFL), Swiss Plasma Center (SPC) , Lausanne CH-1015, Switzerland; Eindhoven University of Technology (TU/e) , Mathematics and Computer Science, Eindhoven, NL-5600MB, NetherlandsÉcole Polytechnique Fédérale de Lausanne (EPFL), Swiss Plasma Center (SPC) , Lausanne CH-1015, SwitzerlandÉcole Polytechnique Fédérale de Lausanne (EPFL), Swiss Plasma Center (SPC) , Lausanne CH-1015, SwitzerlandÉcole Polytechnique Fédérale de Lausanne (EPFL), Swiss Plasma Center (SPC) , Lausanne CH-1015, SwitzerlandEindhoven University of Technology (TU/e) , Mathematics and Computer Science, Eindhoven, NL-5600MB, NetherlandsMaximizing fusion performance in tokamaks relies on high energy confinement, often achieved through distinct operating regimes. The automated labeling of these confinement states is crucial to enable large-scale analyses or for real-time control applications. While this task becomes difficult to automate near state transitions or in marginal scenarios, much success has been achieved with data-driven models. However, these methods generally provide predictions as point estimates, and cannot adequately deal with missing and/or broken input signals. To enable wide-range applicability, we develop methods for confinement state classification with uncertainty quantification and model robustness . We focus on off-line analysis for TCV discharges, distinguishing L-mode, H-mode, and an in-between dithering phase (D). We propose ensembling data-driven methods on two axes: model formulations and feature sets. The former considers a dynamic formulation based on a recurrent Fourier neural operator-architecture and a static formulation based on gradient-boosted decision trees. These models are trained using multiple feature groupings categorized by diagnostic system or physical quantity. A dataset of 302 TCV discharges is fully labeled, and we release it publicly to encourage the community to build upon this work. We evaluate our method quantitatively using Cohen’s kappa coefficient for predictive performance and the expected calibration error for the uncertainty calibration. Furthermore, we discuss performance using a variety of common and alternative scenarios, the performance of individual components, out-of-distribution performance, cases of broken or missing signals, and evaluate conditionally-averaged behavior around different state transitions. Overall, the proposed method can distinguish L, D and H-mode with high performance, can cope with missing or broken signals, and provides meaningful uncertainty estimates.https://doi.org/10.1088/1741-4326/adf349confinement stateH-modeL-modeclassificationmachine learninguncertainty quantification
spellingShingle Yoeri Poels
Cristina Venturini
Alessandro Pau
Olivier Sauter
Vlado Menkovski
the TCV Team
the WPTE Team
Robust confinement state classification with uncertainty quantification through ensembled data-driven methods
Nuclear Fusion
confinement state
H-mode
L-mode
classification
machine learning
uncertainty quantification
title Robust confinement state classification with uncertainty quantification through ensembled data-driven methods
title_full Robust confinement state classification with uncertainty quantification through ensembled data-driven methods
title_fullStr Robust confinement state classification with uncertainty quantification through ensembled data-driven methods
title_full_unstemmed Robust confinement state classification with uncertainty quantification through ensembled data-driven methods
title_short Robust confinement state classification with uncertainty quantification through ensembled data-driven methods
title_sort robust confinement state classification with uncertainty quantification through ensembled data driven methods
topic confinement state
H-mode
L-mode
classification
machine learning
uncertainty quantification
url https://doi.org/10.1088/1741-4326/adf349
work_keys_str_mv AT yoeripoels robustconfinementstateclassificationwithuncertaintyquantificationthroughensembleddatadrivenmethods
AT cristinaventurini robustconfinementstateclassificationwithuncertaintyquantificationthroughensembleddatadrivenmethods
AT alessandropau robustconfinementstateclassificationwithuncertaintyquantificationthroughensembleddatadrivenmethods
AT oliviersauter robustconfinementstateclassificationwithuncertaintyquantificationthroughensembleddatadrivenmethods
AT vladomenkovski robustconfinementstateclassificationwithuncertaintyquantificationthroughensembleddatadrivenmethods
AT thetcvteam robustconfinementstateclassificationwithuncertaintyquantificationthroughensembleddatadrivenmethods
AT thewpteteam robustconfinementstateclassificationwithuncertaintyquantificationthroughensembleddatadrivenmethods