PCS-Controlled Pellet Injection and ITER-Like Gas Puffing to Enable Advanced Density Control Developments on KSTAR

E. Schuster, S. T. Paruchuri, H. Al Khawalde, V. Graber, A. Pajares, J.-W. Juhn, S.-H. Park

4th International Fusion and Plasma Conference (iFPC 2025)

Daejeon, South Korea, June 9-13, 2025

The effectiveness of fueling by gas puffing is expected to be limited in ITER due to the increased difficulty for neutral gas particles to penetrate the hot plasma core, which exhibits edge temperatures of the order of 1 keV. To overcome this limitation, ITER will rely on the injection of frozen fuel pellets, which can penetrate deeper into the plasma core, as the main fueling mechanism. However, this fueling approach, largely unexplored in present devices, presents unique challenges, such as size and velocity effects on the spatiotemporal plasma dynamics, the discrete-time nature of the actuation, and the ablation and penetration dependence on plasma conditions. KSTAR is developing robust density control techniques applicable to ITER and future fusion reactors, which are also expected to significantly impact the exploration of advanced scenarios on KSTAR. The development strategy combines plasma-response modeling, advanced model-based control design, and control testing in an ITER-emulated environment. Recent experimental campaigns on KSTAR utilized the recent commissioning of a pellet injection system capable of injecting deuterium pellets (2 mm diameter, 1.5–2 mm length) at velocities above 200 m/s with an injection frequency of 1–20 Hz. During these experiments, the Pellet Injection System (PIS) was connected for the first time with the Plasma Control System (PCS). This newly developed PCS-controlled pellet-injection capability enables the implementation of feedback-control algorithms for pellet-based density regulation. As a first step, recent experiments demonstrated closed-loop tracking of a density target using a simple non-model-based proportional controller, with performance typical of such algorithms. By modifying the injection frequency as a function of the tracking error, successful tracking was achieved. Additionally, an algorithm called NERF-mode, which emulates ITER-like gas-puffing actuator delays and lags, was implemented and tested. This algorithm enables future tests of density-control algorithms in an ITER-like fueling environment. Data characterizing plasma responses to both pellet injection and NERF-mode gas puffing was collected for model development, validation, and subsequent model-based control design. The detailed experimental results, ongoing work, and future plans will be presented. These plans include the development of 0D and 1D response models, design of model-based adaptive control algorithms, development of density-profile control and observation strategies, integrated gas puffing and pellet injection control, testing in the ITER-like fueling environment on KSTAR using the NERF-mode, and cross-machine validation before extrapolation to ITER.

[1] S.-H. Park et al., Fusion Engineering and Design 146 (2019) 2430–2433.

This work was partially supported by the US Department of Energy under awards DE-SC0010537, DE-FC02-04ER54698 and by the “KSTAR Experimental Collaboration and Fusion Plasma Research (EN2401-12)” through the Korea Institute of Fusion Energy (KFE).