arXiv subjects

Semin Joung

Publications and source records attributed to Semin Joung.

6 recordsLinked to original sources

arXiv2026

FPGA-Accelerated Real-Time Diagnostics at DIII-D Using the SLAC Neural Network Library for ML Inference

In this work, we demonstrate the deployment of a hardware-accelerated machine learning (ML) inference system integrated into a real-time processing at the DIII-D tokamak fusion reactor. The team has successfully deployed an AMD/Xilinx KCU1500 field-programmable gate array (FPGA) into the realtime Plasma Control System (PCS) nodes that receives the live Beam Emission Spectroscopy (BES) signal used for Edge Localized Mode (ELM) forecasting. The FPGA hosts a dense neural network using the SLAC Neural Network Library (SNL) that has been trained to infer the likelihood of disruptive ELM conditions. This likelihood then feeds a separate plasma controller that uses Resonant Magnetic Perturbation coils to suppress the predicted disruptive condition. The SNL allows for on-the-fly updates of the neural network weights and biases without requiring full hardware resynthesis for the FPGA. Judicious design of the neural-network architecture can further allow for the hot-swapping of multiple classification tasks to be executed on the single FPGA, significantly enhancing the real-time adaptability of the system for context-aware control strategies that respond in real-time to evolving reactor conditions. These adaptive weights naturally support continuous model refinement and seamless task switching during live experimental operation. This use case is chosen as a high rate signal processing example that can serve as a template for general ML-based reactor diagnostic processing for active reactor control systems. We see this as an essential development for achieving reactor relevant operation in future continuous operation fusion devices.

physics.plasm-ph↗

arXiv2025

FPGA-Accelerated Real-Time Beam Emission Spectroscopy Diagnostics at DIII-D Using the SLAC Neural Network Library for ML Inference

Achieving reliable real-time control of tokamak plasmas is essential for sustaining high-performance operation in next-generation fusion reactors. A major challenge is the accurate and timely prediction of edge-localized modes (ELMs), especially in high-confinement regimes such as wide-pedestal quiescent H-mode. We present a hardware-accelerated machine learning (ML) inference system integrated into the RTSTAB processing node of the DIII-D real-time diagnostic and control infrastructure. The system uses an AMD/Xilinx KCU1500 FPGA to enable ultra low latency plasma state classification and ELM forecasting. Input features come from real-time Beam Emission Spectroscopy (BES), and the ML model is implemented as a dense neural network using the SLAC Neural Network Library (SNL). A key capability is SNL dynamic parameter loading, which allows on-the-fly updates of neural network weights and biases without hardware resynthesis. This enables multiple classification tasks on a single FPGA design and supports adaptive control strategies that respond to evolving plasma conditions. By decoupling inference from fixed-weight configurations, the system supports continuous model refinement and seamless task switching during live operation. The SNL-based inference engine is fully integrated with the FPGA in the DIII-D RTSTAB Plasma Control System (PCS), improving ELM avoidance, confinement, and operational stability. These results show the feasibility of embedding dynamically reconfigurable FPGA-based ML inference into real-time fusion diagnostic pipelines, providing a scalable and resilient path toward intelligent and autonomous plasma control in future magnetic confinement fusion devices.

physics.plasm-ph↗

arXiv2023

Bayesian neural network for plasma equilibria in the Korea Superconducting Tokamak Advanced Research

Fusion-graded plasmas are one of the physically complex systems, resulting in continuous establishment of plasma theories for unclarified physical phenomena in order to thoroughly control nuclear fusion reactors. Deep learning has drawn vast attention to this field of controlled fusion plasma to link physical phenomena with control-relevant parameters without a deepened understanding about plasma theories. Albeit, quantifying the uncertainty of deep learning models has been constantly requested due to their fundamental shortage of physical understanding. Thus, a concept of a reliable deep learning model to be able to present their probability distributions is raised as well as a method to inculcate physical theories in the model is also concerned.

physics.plasm-ph↗

arXiv2022

Automatic Identification of Edge Localized Modes in the DIII-D Tokamak

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 waste divertor. Machine learning-based prediction and control may provide for online mitigation of these damaging modes before they grow too large to suppress. To that end, large labeled datasets are required for 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.

physics.plasm-ph↗

arXiv2019

Deep neural network Grad-Shafranov solver constrained with measured magnetic signals

A neural network solving Grad-Shafranov equation constrained with measured magnetic signals to reconstruct magnetic equilibria in real time is developed. Database created to optimize the neural network's free parameters contain off-line EFIT results as the output of the network from $1,118$ KSTAR experimental discharges of two different campaigns. Input data to the network constitute magnetic signals measured by a Rogowski coil (plasma current), magnetic pick-up coils (normal and tangential components of magnetic fields) and flux loops (poloidal magnetic fluxes). The developed neural networks fully reconstruct not only the poloidal flux function $ψ\left( R, Z\right)$ but also the toroidal current density function $j_ϕ\left( R, Z\right)$ with the off-line EFIT quality. To preserve robustness of the networks against a few missing input data, an imputation scheme is utilized to eliminate the required additional training sets with large number of possible combinations of the missing inputs.

physics.plasm-ph↗

arXiv2018

Bayesian with Gaussian process based missing input imputation scheme for reconstructing magnetic equilibria in real time

A Bayesian with GP(Gaussian Process)-based numerical method to impute a few missing magnetic signals caused by impaired magnetic probes during tokamak operations is developed such that the real-time reconstruction of magnetic equilibria, whose performance strongly depends on the measured magnetic signals and their intactness, are affected minimally. Likelihood of the Bayesian model constructed with the Maxwell's equations, specifically Gauss's law of magnetism and Ampère's law, results in infinite number of solutions if two or more magnetic signals are missing. This undesirable characteristic of the Bayesian model is remediated by coupling the model with the Gaussian process. Our proposed numerical method infers the missing magnetic signals correctly in less than $1$\:msec suitable for real-time reconstruction of magnetic equilibria during tokamak operations. The method can also be used for a neural network that reconstructs magnetic equilibria trained with a complete set of magnetic signals. Without our proposed imputation method, such a neural network would become useless if missing signals are not tolerable by the network.

physics.plasm-ph↗