Development of machine learning model for automatic ELM-burst detection without hyperparameter adjustment in KSTAR tokamak

Song, Jiheon; Joung, Semin; Ghim, Young-chul; Hahn, S.H.; Jang, Jin; Lee, Jungpyo

doi:10.1016/j.net.2022.08.026

Cited by 4 publications

(3 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Applications of deep learning [36] in the field of observing or predicting plasma instabilities are an active research area, such as observing ELMs [32][33][34] and Alfvén eigenmode (AE mode) [47], or predicting tearing mode instabilities [48,49] and plasma disruptions [50]. Therefore, we propose a neural network, BES-ELMnet, which can predict the type-I ELM onsets [5], i.e.…”

Section: Introductionmentioning

confidence: 99%

“…The ELM onsets can be directly identified, such as using signal peaks in the measured D α line and fluctuating magnetic fields from the visible D α filterscope [28] and the Mirnov coil [29], respectively, and sudden intensity changes in the lineaveraged electron density n e measured from the interferometer [30] along a line of sight through the edge plasma. Hence, attempts have been made to detect the ELMs automatically [31][32][33][34][35] based on deep neural networks [32][33][34]36], sequential probability ratio test [31,37] and a search algorithm [35,38]. These previous works can be used in real time [22,33,35] or at least have potential real-time availability [31,32,34,35].…”

Section: Introductionmentioning

confidence: 99%

“…Hence, attempts have been made to detect the ELMs automatically [31][32][33][34][35] based on deep neural networks [32][33][34]36], sequential probability ratio test [31,37] and a search algorithm [35,38]. These previous works can be used in real time [22,33,35] or at least have potential real-time availability [31,32,34,35]. However, they only observe the ELMs after the bursts occur; i.e.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Tokamak edge localized mode onset prediction with deep neural network and pedestal turbulence

Joung,

Smith,

McKee

et al. 2024

Nucl. Fusion

Self Cite

View full text Add to dashboard Cite

A neural network, BES-ELMnet, predicting a quasi-periodic disruptive eruption of the plasma energy and particles known as edge localized mode (ELM) onset is developed with observed pedestal turbulence from the beam emission spectroscopy system in D\rom{3}-D. BES-ELMnet has convolutional and fully-connected layers, taking two-dimensional plasma fluctuations with a temporal window of size 128\:$\mu$s and generating a scalar output which can be interpreted as a probability of the upcoming ELM onset. As approximately labelled inter-ELM broadband (15\:kHz\:$\leq f\leq$\:150\:kHz) fluctuations are given to the network, BES-ELMnet learns by itself ELM-related precursors arising before the onsets through supervised learning scheme. BES-ELMnet achieves the gradually increasing ELM onset probabilities between two consecutive ELMs during the inter-ELM phases and can forecast the first ELM onsets which occur after the high confinement mode transition. We further investigate the network generality in terms of the selected frequency band to ensure the use of BES-ELMnet for various operation regimes without changing the trained architecture. Therefore, our novel prediction method will enhance a proactive high confinement mode control of fusion-grade plasmas.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Tokamak edge localized mode onset prediction with deep neural network and pedestal turbulence

Joung,

Smith,

McKee

et al. 2024

Nucl. Fusion

Self Cite

View full text Add to dashboard Cite

show abstract

Automatic identification of edge localized modes in the DIII-D tokamak

O'Shea

Joung

Smith

et al. 2023

APL Machine Learning

View full text Add to dashboard Cite

Fusion power production in tokamaks uses discharge configurations that risk producing strong type I edge localized modes. The largest of these modes will likely increase impurities in the plasma and potentially damage plasma facing components, such as the protective heat and particle divertor. Machine learning-based prediction and control may provide for the automatic detection and mitigation of these damaging modes before they grow too large to suppress. To that end, large labeled datasets are required for the supervised training of machine learning models. We present an algorithm that achieves 97.7% precision when automatically labeling edge localized modes in the large DIII-D tokamak discharge database. The algorithm has no user controlled parameters and is largely robust to tokamak and plasma configuration changes. This automatically labeled database of events can subsequently feed future training of machine learning models aimed at autonomous edge localized mode control and suppression.

show abstract

Low latency optical-based mode tracking with machine learning deployed on FPGAs on a tokamak

Wei,

Forelli,

Hansen

et al. 2024

Review of Scientific Instruments

View full text Add to dashboard Cite

Active feedback control in magnetic confinement fusion devices is desirable to mitigate plasma instabilities and enable robust operation. Optical high-speed cameras provide a powerful, non-invasive diagnostic and can be suitable for these applications. In this study, we process high-speed camera data, at rates exceeding 100 kfps, on in situ field-programmable gate array (FPGA) hardware to track magnetohydrodynamic (MHD) mode evolution and generate control signals in real time. Our system utilizes a convolutional neural network (CNN) model, which predicts the n = 1 MHD mode amplitude and phase using camera images with better accuracy than other tested non-deep-learning-based methods. By implementing this model directly within the standard FPGA readout hardware of the high-speed camera diagnostic, our mode tracking system achieves a total trigger-to-output latency of 17.6 μs and a throughput of up to 120 kfps. This study at the High Beta Tokamak-Extended Pulse (HBT-EP) experiment demonstrates an FPGA-based high-speed camera data acquisition and processing system, enabling application in real-time machine-learning-based tokamak diagnostic and control as well as potential applications in other scientific domains.

show abstract

Development of machine learning model for automatic ELM-burst detection without hyperparameter adjustment in KSTAR tokamak

Cited by 4 publications

References 30 publications

Tokamak edge localized mode onset prediction with deep neural network and pedestal turbulence

Tokamak edge localized mode onset prediction with deep neural network and pedestal turbulence

Automatic identification of edge localized modes in the DIII-D tokamak

Low latency optical-based mode tracking with machine learning deployed on FPGAs on a tokamak

Contact Info

Product

Resources

About