Augmenting the extrapolation capability of disruption prediction to extended parameter regimes by predict-first neural network

Abstract The prediction, mitigation, and avoidance of disruptions are critical prerequisites for ensuring the safe operation of future fusion reactors. While machine learning-based disruption prediction techniques have achieved high accuracy in recent years, their application to a new device demands strong extrapolation capability across parameter regimes. This capability is essential to maintain robust algorithm performance as the operational space of the device expands, while it is also a main shortcoming for most machine learning techniques. This study evaluates the extrapolation capability of a disruption prediction algorithm using multiple subsets of experimental data from HL-3 tokamak, divided by key plasma parameters, yielding three key findings. Firstly, the area under the receiver-operator characteristic curve (AUC) of a standard deep learning algorithm trained on the subset with low parameters degrades to around 0.750, 0.666, and 0.540 when extrapolated to three progressively distant extrapolation regimes. However, the Predict-First Neural Network (PFNN) maintains an AUC of about 0.873, 0.823 and 0.701. Secondly, the algorithm's accuracy shows a clear contrast in degradation behavior: when extrapolating along electromagnetic parameters like plasma current, performance declines slowly and continuously; when extrapolating along energy-related parameters like stored energy, performance remains initially stable but deteriorates abruptly beyond a certain threshold. Finally, after rigorously accounting for the impact of parameter regime extrapolation and adjusting the algorithm deployment strategy, deep learning proved effective in protecting the device during HL-3's high-current and high-beta commissioning phases. This study provides valuable experience for implementing disruption prediction algorithms on new devices and is expected to offer practical reference for the initial operation of future fusion facilities like ITER.

Authors

Institutions

Publication Details

Journal
Nuclear Fusion
Published
2026-11-17
DOI
https://doi.org/10.1088/1741-4326/aec306
Primary Topic
Magnetic confinement fusion research
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Augmenting the extrapolation capability of disruption prediction to extended parameter regimes by predict-first neural network

W.L. Zhong, T.F. Sun, Xia Fan, F.J. Artola et al.
2 links
Nuclear Fusion
Magnetic confinement fusion research
article

Augmenting the extrapolation capability of disruption prediction to extended parameter regimes by predict-first neural network

W.L. Zhong, T.F. Sun, Xia Fan, F.J. Artola, 张轶泼, Junzhao zhang, Jiyuan Li, Yihang Chen, Zongyu Yang, Zhaohe Xu, Ruoshu Qiu, B Li, Da Li, Y B Dong
article en

Abstract

Abstract The prediction, mitigation, and avoidance of disruptions are critical prerequisites for ensuring the safe operation of future fusion reactors. While machine learning-based disruption prediction techniques have achieved high accuracy in recent years, their application to a new device demands strong extrapolation capability across parameter regimes. This capability is essential to maintain robust algorithm performance as the operational space of the device expands, while it is also a main shortcoming for most machine learning techniques. This study evaluates the extrapolation capability of a disruption prediction algorithm using multiple subsets of experimental data from HL-3 tokamak, divided by key plasma parameters, yielding three key findings. Firstly, the area under the receiver-operator characteristic curve (AUC) of a standard deep learning algorithm trained on the subset with low parameters degrades to around 0.750, 0.666, and 0.540 when extrapolated to three progressively distant extrapolation regimes. However, the Predict-First Neural Network (PFNN) maintains an AUC of about 0.873, 0.823 and 0.701. Secondly, the algorithm's accuracy shows a clear contrast in degradation behavior: when extrapolating along electromagnetic parameters like plasma current, performance declines slowly and continuously; when extrapolating along energy-related parameters like stored energy, performance remains initially stable but deteriorates abruptly beyond a certain threshold. Finally, after rigorously accounting for the impact of parameter regime extrapolation and adjusting the algorithm deployment strategy, deep learning proved effective in protecting the device during HL-3's high-current and high-beta commissioning phases. This study provides valuable experience for implementing disruption prediction algorithms on new devices and is expected to offer practical reference for the initial operation of future fusion facilities like ITER.

Nuclear Fusion
ITER (FR), Southwestern Institute of Physics (CN)
National Natural Science Foundation of China, Natural Science Foundation of Sichuan Province
Magnetic confinement fusion research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon