Boseong Kim
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Boseong Kim has a total of 62 co-authors. Here are the top co-authors by number of shared publications:
- Yong-Su Na
Department of Nuclear Engineering, Seoul National University 1 , Seoul 08826
4 shared publications - Youngho Lee
Korea Institute of Fusion Energy
4 shared publications - Sang-Kyeun Kim
Princeton Plasma Physics Laboratory
3 shared publications - Chan-Young Lee
Korea Institute of Fusion Energy, Daejeon, Korea, Republic Of
2 shared publications - Hyung Ho Lee
Korea Institute of Fusion Energy
2 shared publications - Jekil Lee
Korea Institute of Fusion Energy
2 shared publications
Latest Publications
- Open Access
- 1 August 2026
- Published by IOP Publishing in Plasma Physics and Controlled Fusion
- Vol. 68 (8) , 085041
Validation of integrated modeling of KSTAR plasmas with carbon divertor using JINTRAC and prediction for tungsten divertor
The JINTRAC integrated modeling suite has been adapted for the KSTAR tokamak, with its primary goal being the establishment of a core-edge integrated modeling framework for predictive analysis of KSTAR plasmas. This framework was validated by benchmarking JINTRAC’s physics modules against both established codes and experimental data from a KSTAR carbon (C) divertor discharge (#25458), demonstrating high fidelity. Based on the validation, predictive core-edge coupled simulations were performed to investigate the effects of changing the divertor material from C to tungsten (W). The simulations predicted a significant increase in core radiation and a corresponding decrease in plasma temperatures, consistent with the behavior of high- Z materials. Specifically, the high tungsten influx causes radiative cooling in the pedestal and core regions, leading to a reduction in edge bootstrap current and a compensatory increase in core inductive current. This redistribution of current density drops the central safety factor ( q 0 ) below unity, triggering sawtooth instability which serves to suppress core impurity accumulation. Sensitivity analysis indicated that while toroidal rotation exhibits a non-monotonic correlation with tungsten profile peaking, the total tungsten inventory and radiated power ( P rad ) increased with higher rotation. Additionally, increasing the separatrix electron density effectively suppressed the tungsten source at the divertor target, yet the reduction in core tungsten inventory was limited compared to the extent of source reduction. These findings imply that achieving high-performance operation requires a comprehensive control strategy that minimizes the absolute impurity influx across the separatrix through combined source suppression and transport control.
- 1 August 2026
- Published by IOP Publishing in Plasma Physics and Controlled Fusion
- Vol. 68 (8) , 085041
Open AccessValidation of integrated modeling of KSTAR plasmas with carbon divertor using JINTRAC and prediction for tungsten divertor
The JINTRAC integrated modeling suite has been adapted for the KSTAR tokamak, with its primary goal being the establishment of a core-edge integrated modeling framework for predictive analysis of KSTAR plasmas. This framework was validated by benchmarking JINTRAC’s physics modules against both established codes and experimental data from a KSTAR carbon (C) divertor discharge (#25458), demonstrating high fidelity. Based on the validation, predictive core-edge coupled simulations were performed to investigate the effects of changing the divertor material from C to tungsten (W). The simulations predicted a significant increase in core radiation and a corresponding decrease in plasma temperatures, consistent with the behavior of high- Z materials. Specifically, the high tungsten influx causes radiative cooling in the pedestal and core regions, leading to a reduction in edge bootstrap current and a compensatory increase in core inductive current. This redistribution of current density drops the central safety factor ( q 0 ) below unity, triggering sawtooth instability which serves to suppress core impurity accumulation. Sensitivity analysis indicated that while toroidal rotation exhibits a non-monotonic correlation with tungsten profile peaking, the total tungsten inventory and radiated power ( P rad ) increased with higher rotation. Additionally, increasing the separatrix electron density effectively suppressed the tungsten source at the divertor target, yet the reduction in core tungsten inventory was limited compared to the extent of source reduction. These findings imply that achieving high-performance operation requires a comprehensive control strategy that minimizes the absolute impurity influx across the separatrix through combined source suppression and transport control.
- Open Access
- 25 May 2026
- Published by Elsevier in SSRN Electronic Journal
Bayesian optimization for design space exploration of tokamak fusion reactors
Designing a tokamak fusion reactor requires exploring a high-dimensional parameter space where multiple coupled variables must simultaneously satisfy physics constraints and engineering limits. Conventional exhaustive search approaches become computationally prohibitive as dimensionality increases. This paper demonstrates the application of multi-objective Bayesian optimization (MOBO) to tokamak design space exploration using a low-fidelity system analysis code.We compare state-of-the-art acquisition functions against an exhaustive search baseline for bi-objective optimization minimizing major radius (a capital cost proxy) and maximizing net electric power. Results show that all MOBO methods recover Pareto fronts comparable to exhaustive search while requiring orders of magnitude fewer function evaluations. We investigate the effects of acquisition function choice, batch size for parallel evaluation, and constraint handling strategies on convergence behavior. The methodology provides practical guidelines for efficient design optimization and establishes a framework applicable to higher-fidelity,computationally expensive plasma simulations.
- 25 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessBayesian optimization for design space exploration of tokamak fusion reactors
Designing a tokamak fusion reactor requires exploring a high-dimensional parameter space where multiple coupled variables must simultaneously satisfy physics constraints and engineering limits. Conventional exhaustive search approaches become computationally prohibitive as dimensionality increases. This paper demonstrates the application of multi-objective Bayesian optimization (MOBO) to tokamak design space exploration using a low-fidelity system analysis code.We compare state-of-the-art acquisition functions against an exhaustive search baseline for bi-objective optimization minimizing major radius (a capital cost proxy) and maximizing net electric power. Results show that all MOBO methods recover Pareto fronts comparable to exhaustive search while requiring orders of magnitude fewer function evaluations. We investigate the effects of acquisition function choice, batch size for parallel evaluation, and constraint handling strategies on convergence behavior. The methodology provides practical guidelines for efficient design optimization and establishes a framework applicable to higher-fidelity,computationally expensive plasma simulations.
- Open Access
- 1 January 2026
- Published by IOP Publishing in Plasma Physics and Controlled Fusion
- Vol. 68 (1) , 015007
Real-time reconstruction and control of pedestal-top electron density using RMP and gas puff at KSTAR
We report the experimental results of controlling the pedestal-top electron density by applying resonant magnetic perturbation (RMP) with in-vessel control coils and the main gas puff in the 2024-2025 KSTAR experimental campaign. The density is reconstructed using a parameterized ψ N grid and five channels of line-averaged density measured by the two-colored interferometer (TCI). The reconstruction procedure is accelerated by deploying a multi-layer perceptron to run in approximately 120 µ s and is sufficiently fast for real-time control. A proportional-integral controller was adopted, with the controller gains estimated from the system identification procedure. The experimental results demonstrate that the developed controller can follow a dynamic target while exclusively using both actuators. The absolute percentage errors between the electron density at ψ N = 0.89 and the target were approximately 1.5% median and a 2.5% average, respectively. The developed controller can even lower the density by using the pump-out mechanism under RMP, and it can follow a more dynamic range of density targets than a single actuator controller. The developed controller will enable experimental scenario exploration within a shot by dynamically setting the density target or maintaining a constant electron density within a discharge.
- 1 January 2026
- Published by IOP Publishing in Plasma Physics and Controlled Fusion
- Vol. 68 (1) , 015007
Open AccessReal-time reconstruction and control of pedestal-top electron density using RMP and gas puff at KSTAR
We report the experimental results of controlling the pedestal-top electron density by applying resonant magnetic perturbation (RMP) with in-vessel control coils and the main gas puff in the 2024-2025 KSTAR experimental campaign. The density is reconstructed using a parameterized ψ N grid and five channels of line-averaged density measured by the two-colored interferometer (TCI). The reconstruction procedure is accelerated by deploying a multi-layer perceptron to run in approximately 120 µ s and is sufficiently fast for real-time control. A proportional-integral controller was adopted, with the controller gains estimated from the system identification procedure. The experimental results demonstrate that the developed controller can follow a dynamic target while exclusively using both actuators. The absolute percentage errors between the electron density at ψ N = 0.89 and the target were approximately 1.5% median and a 2.5% average, respectively. The developed controller can even lower the density by using the pump-out mechanism under RMP, and it can follow a more dynamic range of density targets than a single actuator controller. The developed controller will enable experimental scenario exploration within a shot by dynamically setting the density target or maintaining a constant electron density within a discharge.
- Open Access
- 13 November 2025
- Published by Elsevier in Fusion Engineering and Design
- Vol. 222
Neural network-based analysis of charge exchange spectra in KSTAR
- 13 November 2025
- Published by Elsevier in Fusion Engineering and Design
- Vol. 222
Open AccessNeural network-based analysis of charge exchange spectra in KSTAR
- 6 May 2025
- Published by IOP Publishing in Plasma Physics and Controlled Fusion
- Vol. 67 (5) , 055037
Development of V-band Doppler reflectometer based on frequency steering phased array antenna
KSTAR Doppler reflectometer (DR) is designed based on frequency steering phased array antenna (FSPAA) operating in X-mode over the full V-band (50–75 GHz). In the FSPAA system, the microwave radiation angle varies as a function of frequency, allowing rapid changes in the radiation angle without mechanical movement. This is a very attractive advantage in fusion reactor environments where the concept of maintenance-free is important. However, FSPAA has limitations in microwave radiation in the broadside direction due to the open-stopband problem. In this paper, various types of series feeds are developed to suppress open-stopband and their performance is tested based on computer simulations. Directional coupling feed shows the best performance. Two prototype FSPAAs consisting of 8 antenna elements with directional coupling feed networks have been manufactured using electroforming technology. The performance of the manufactured FSPAAs are experimentally measured. Since microwaves are radiated with uniform power over an angular range of ± 40 ∘ for a frequency sweep of 3 GHz, the angular scan is repeated 9 times over the entire V-band. A prototype KSTAR DR based on a bistatic antenna configuration is assembled using the two FSPAAs. The performance of the prototype DR is tested using a rotating corrugated aluminum wheel in the laboratory.
- 6 May 2025
- Published by IOP Publishing in Plasma Physics and Controlled Fusion
- Vol. 67 (5) , 055037
Development of V-band Doppler reflectometer based on frequency steering phased array antenna
KSTAR Doppler reflectometer (DR) is designed based on frequency steering phased array antenna (FSPAA) operating in X-mode over the full V-band (50–75 GHz). In the FSPAA system, the microwave radiation angle varies as a function of frequency, allowing rapid changes in the radiation angle without mechanical movement. This is a very attractive advantage in fusion reactor environments where the concept of maintenance-free is important. However, FSPAA has limitations in microwave radiation in the broadside direction due to the open-stopband problem. In this paper, various types of series feeds are developed to suppress open-stopband and their performance is tested based on computer simulations. Directional coupling feed shows the best performance. Two prototype FSPAAs consisting of 8 antenna elements with directional coupling feed networks have been manufactured using electroforming technology. The performance of the manufactured FSPAAs are experimentally measured. Since microwaves are radiated with uniform power over an angular range of ± 40 ∘ for a frequency sweep of 3 GHz, the angular scan is repeated 9 times over the entire V-band. A prototype KSTAR DR based on a bistatic antenna configuration is assembled using the two FSPAAs. The performance of the prototype DR is tested using a rotating corrugated aluminum wheel in the laboratory.
- Open Access
- 1 January 2025
- Published by AIP Publishing in Physics of Plasmas
- Vol. 32 (1)
Observation of edge kink-like modes induced by resonant magnetic perturbations in KSTAR plasmas and their effects on density pump-out
In tokamaks, it is commonly observed that the application of resonant magnetic perturbations (RMPs) leads to a reduction in plasma density. In this study, we show that this decrease in density is accompanied by kink-like modes in the plasma edge region in KSTAR. The dynamics of these modes is observed in the toroidal and poloidal directions using multiple diagnostics. It is captured that the phase of the edge kink-like modes aligns with the phase of the applied RMPs. In particular, a nonuniform plasma surface displacement due to these modes is measured along the poloidal direction using a novel image processing technique on in-vessel TV data. The symmetry-breaking effect of the displacement is known to be much larger than that of the applied RMPs. Thus, the modification in the magnetic field strength B on the distorted surface due to the displacement can lead to significant enhancement of the neoclassical particle transport. In this study, we calculate the enhanced neoclassical electron particle flux using the experimentally estimated variation of B in the presence of the edge kink-like modes. Transport analysis shows that the enhanced particle transport caused by the broken symmetry in the presence of the edge kink-like modes can account for a significant portion of the observed density pump-out by RMPs.
- 1 January 2025
- Published by AIP Publishing in Physics of Plasmas
- Vol. 32 (1)
Open AccessObservation of edge kink-like modes induced by resonant magnetic perturbations in KSTAR plasmas and their effects on density pump-out
In tokamaks, it is commonly observed that the application of resonant magnetic perturbations (RMPs) leads to a reduction in plasma density. In this study, we show that this decrease in density is accompanied by kink-like modes in the plasma edge region in KSTAR. The dynamics of these modes is observed in the toroidal and poloidal directions using multiple diagnostics. It is captured that the phase of the edge kink-like modes aligns with the phase of the applied RMPs. In particular, a nonuniform plasma surface displacement due to these modes is measured along the poloidal direction using a novel image processing technique on in-vessel TV data. The symmetry-breaking effect of the displacement is known to be much larger than that of the applied RMPs. Thus, the modification in the magnetic field strength B on the distorted surface due to the displacement can lead to significant enhancement of the neoclassical particle transport. In this study, we calculate the enhanced neoclassical electron particle flux using the experimentally estimated variation of B in the presence of the edge kink-like modes. Transport analysis shows that the enhanced particle transport caused by the broken symmetry in the presence of the edge kink-like modes can account for a significant portion of the observed density pump-out by RMPs.
- 1 October 2024
- Published by AIP Publishing in Review of Scientific Instruments
- Vol. 95 (10)
Classification of the L-, H-mode, and plasma-free state: Convolutional neural networks and variational autoencoders on the edge reflectometer for KSTAR
Classifying and monitoring the L-, H-mode, and plasma-free state are essential for the stable operational control of tokamaks. Edge reflectometry measures plasma density profiles, but the large volume of data and complexity in reconstruction pose significant challenges. There is a need for efficient methods to analyze complex reflectometer data in real-time, which can be addressed using advanced computational techniques. Here, we show that machine learning (ML) techniques can classify discharge states using raw signal data from an edge reflectometer installed on the Korea Superconducting Tokamak Advanced Research. The deep convolutional neural network models achieved classification accuracy of up to 99% when using 2D spectrogram inputs, demonstrating a significant improvement over 1D raw signal inputs. Additionally, the variational autoencoder model effectively clustered the discharge states in the latent space without any label information, further validating the model’s capability to classify discharge states. These results suggest that the ML model can effectively handle the complexity of reflectometer data and accurately classify plasma discharge states. This approach not only facilitates real-time diagnosis but also reduces the need for manual data processing.
- 1 October 2024
- Published by AIP Publishing in Review of Scientific Instruments
- Vol. 95 (10)
Classification of the L-, H-mode, and plasma-free state: Convolutional neural networks and variational autoencoders on the edge reflectometer for KSTAR
Classifying and monitoring the L-, H-mode, and plasma-free state are essential for the stable operational control of tokamaks. Edge reflectometry measures plasma density profiles, but the large volume of data and complexity in reconstruction pose significant challenges. There is a need for efficient methods to analyze complex reflectometer data in real-time, which can be addressed using advanced computational techniques. Here, we show that machine learning (ML) techniques can classify discharge states using raw signal data from an edge reflectometer installed on the Korea Superconducting Tokamak Advanced Research. The deep convolutional neural network models achieved classification accuracy of up to 99% when using 2D spectrogram inputs, demonstrating a significant improvement over 1D raw signal inputs. Additionally, the variational autoencoder model effectively clustered the discharge states in the latent space without any label information, further validating the model’s capability to classify discharge states. These results suggest that the ML model can effectively handle the complexity of reflectometer data and accurately classify plasma discharge states. This approach not only facilitates real-time diagnosis but also reduces the need for manual data processing.
- Open Access
- 4 June 2024
- Published by IOP Publishing in Nuclear Fusion
- Vol. 64 (7) , 076031
Development of a neural network model for peeling–ballooning stability analysis in the KSTAR tokamak pedestals
The neural network model, MISHKA-NN is developed to mitigate the computational burden associated with the linear ideal magnetohydrodynamic (MHD) stability analysis of the pedestal based on the peeling–ballooning (P–B) model. By utilizing both 1D plasma profiles (current density, pressure gradient, and safety factor) and 0D parameters (plasma geometry, total current, and toroidal mode number), the model predicts linear growth rate of edge-localized ideal MHD instability in a given equilibrium state. By enabling the prediction of each instability within a second, the model reduces the time required for plotting a pedestal P–B stability diagram ( j − α diagram) from approximately 100 CPU hours to a few CPU minutes. Notably, even with the utilization of parametric pressure and current profiles and plasma boundary shapes for the training dataset, the model shows a satisfactory level of performance in benchmarking the j − α diagram for the reconstructed equilibrium from a KSTAR tokamak experiment. We anticipate the model to serve as a versatile alternative to 2D linear MHD stability codes, alleviating numerical costs.
- 4 June 2024
- Published by IOP Publishing in Nuclear Fusion
- Vol. 64 (7) , 076031
Open AccessDevelopment of a neural network model for peeling–ballooning stability analysis in the KSTAR tokamak pedestals
The neural network model, MISHKA-NN is developed to mitigate the computational burden associated with the linear ideal magnetohydrodynamic (MHD) stability analysis of the pedestal based on the peeling–ballooning (P–B) model. By utilizing both 1D plasma profiles (current density, pressure gradient, and safety factor) and 0D parameters (plasma geometry, total current, and toroidal mode number), the model predicts linear growth rate of edge-localized ideal MHD instability in a given equilibrium state. By enabling the prediction of each instability within a second, the model reduces the time required for plotting a pedestal P–B stability diagram ( j − α diagram) from approximately 100 CPU hours to a few CPU minutes. Notably, even with the utilization of parametric pressure and current profiles and plasma boundary shapes for the training dataset, the model shows a satisfactory level of performance in benchmarking the j − α diagram for the reconstructed equilibrium from a KSTAR tokamak experiment. We anticipate the model to serve as a versatile alternative to 2D linear MHD stability codes, alleviating numerical costs.
- Open Access
- 25 September 2023
- Published by IOP Publishing in Nuclear Fusion
- Vol. 63 (12) , 126013
Investigation of performance enhancement by balanced double-null shaping in KSTAR
We report experimental observations on the effect of plasma boundary shaping towards balanced double-null (DN) configuration on the plasma performance in KSTAR. The transition from a single-null to a DN configuration resulted in improved plasma performance, manifested through changes in the pedestal region, decreased density, and core MHD activity variation. Specifically, the DN transition led to a wider and higher pedestal structure, accompanied by grassy edge-localized modes (ELMs) characteristics. The density decrease was a prerequisite for performance enhancement during DN shaping, increasing fast ion confinement. Optimizing the plasma near the core region was associated with the suppression of sawtooth instabilities and the occurrence of fishbone modes during the DN transition. Integrated modeling demonstrated that secondary effects of the DN shaping could increase core thermal energy confinement.
- 25 September 2023
- Published by IOP Publishing in Nuclear Fusion
- Vol. 63 (12) , 126013
Open AccessInvestigation of performance enhancement by balanced double-null shaping in KSTAR
We report experimental observations on the effect of plasma boundary shaping towards balanced double-null (DN) configuration on the plasma performance in KSTAR. The transition from a single-null to a DN configuration resulted in improved plasma performance, manifested through changes in the pedestal region, decreased density, and core MHD activity variation. Specifically, the DN transition led to a wider and higher pedestal structure, accompanied by grassy edge-localized modes (ELMs) characteristics. The density decrease was a prerequisite for performance enhancement during DN shaping, increasing fast ion confinement. Optimizing the plasma near the core region was associated with the suppression of sawtooth instabilities and the occurrence of fishbone modes during the DN transition. Integrated modeling demonstrated that secondary effects of the DN shaping could increase core thermal energy confinement.