Integration and evolution of architectural components in the ITER plasma control system: deployment and experimental validation at KSTAR

Abstract

The ITER Plasma Control System (PCS) is a large-scale, highly integrated control system designed to autonomously execute all phases of tokamak pulse operation. Given its extended development horizon, strong functional couplings, and stringent reliability requirements, PCS development follows a systematic methodology encompassing systems engineering, model-based design, and staged deployment. Building on an initial case study of the Real-Time Framework (RTF) applied to the ITER PCS, this paper details the improvements and functional extensions introduced in subsequent development stages, demonstrating that the system has achieved the stability and flexibility required for real tokamak operation. Key structural components — including pulse supervision control with integrated exception handling, the segment-based decision-making process that governs the pulse trajectory, and adaptive gain scheduling for plasma controllers — are elaborated in detail. The PCS Provisioning System (PPS), which orchestrates end-to-end PCS operation spanning configuration preparation, pulse execution, and post-pulse analysis, is presented as the central automation layer. The paper further presents the deployment of the ITER PCS architecture on the operating tokamak KSTAR as a practical validation case, describing the methodology for adapting the machine-independent core architecture to an external device and reporting the results of the 2026 experimental campaign.