• Corpus ID: 246063858

Machine-Learning enabled analysis of ELM filament dynamics in KSTAR

@article{Jacobus2022MachineLearningEA,
  title={Machine-Learning enabled analysis of ELM filament dynamics in KSTAR},
  author={Cooper Jacobus and Minjun Choi and Ralph Kube},
  journal={ArXiv},
  year={2022},
  volume={abs/2201.07941}
}
The emergence and dynamics of filamentary structures associated with edge-localized modes (ELMs) inside tokamak plasmas during high-confinement mode is regularly studied using Electron Cyclotron Emission Imaging (ECEI) diagnostic systems. ECEI allows to infer electron temperature variations, often across a poloidal cross-section. Previously, detailed analyses of filamentary dynamics and classification of the precursors to ELM crashes have been done manually. We present a machine-learning-based… 

References

SHOWING 1-10 OF 71 REFERENCES
A machine learning approach to identify the universality of solitary perturbations accompanying boundary bursts in magnetized toroidal plasmas
TLDR
It is confirmed that the large burst at the plasma boundary that can fatally damage the fusion device always involves the emergence of SP, suggesting that the SP is a key to understanding and controlling of the boundary burst in magnetized toroidal plasmas.
Two-dimensional imaging of edge-localized modes in KSTAR plasmas unperturbed and perturbed by n=1 external magnetic fields
The temporal evolution of edge-localized modes (ELMs) has been studied using a 2-D electron cyclotron emission imaging system in the KSTAR tokamak. The ELMs are observed to evolve in three
Development of KSTAR ECE imaging system for measurement of temperature fluctuations and edge density fluctuations.
TLDR
An improved ECEI system capable of visualizing both high- and low-field sides simultaneously with considerably better spatial coverage has been developed for the KSTAR tokamak in order to capture the full picture of core MHD dynamics.
Progress Toward Interpretable Machine Learning–Based Disruption Predictors Across Tokamaks
TLDR
This paper defines physics-based indicators of disruption precursors based on temperature, density, and radiation profiles that are currently not used in many other machine learning predictors for D III-D data to well-describe impurity accumulation events in both DIII-D and JET discharges that eventually disrupt.
Disruption prediction at JET through deep convolutional neural networks using spatiotemporal information from plasma profiles
TLDR
A deep-convolutional neural network is proposed to provide early detection of disruptive events at JET, and the CNN ability to learn relevant features, avoiding hand-engineered feature extraction, has been exploited to extract the spatiotemporal information from 1D plasma profiles.
Deep learning based surrogate models for first-principles global simulations of fusion plasmas
TLDR
A workflow has been presented to develop deep-learning based surrogate models for the first-principle simulations using the gyrokinetic toroidal code (GTC) to develop physics-based fast instability simulators that fits the requirement of the real-time plasma control system.
2D ECE measurements of type-I edge localized modes at ASDEX Upgrade
The installation of a 2D electron cyclotron emission imaging (ECEI) diagnostic on ASDEX Upgrade has provided a new means to observe the nature of edge localized modes (ELMs). For a series of ELMs in
Confinement and ELM characteristics of H-mode plasmas in KSTAR
The latest results of confinement and edge-localized mode (ELM) characteristics of Korea Superconducting Tokamak Advanced Research (KSTAR) H-mode plasmas are reported. The estimation of fast ion
Primary edge localized mode filament structure in the National Spherical Torus Experimenta)
Edge localized modes (ELMs) are routinely seen in the National Spherical Torus Experiment (NSTX) [M. Ono, Nucl. Fusion 40, 557 (2000)]. These unstable modes give rise to plasma filaments that burst
A review of ELMs in divertor tokamaks
...
...