Bridging the gap: Deploying AI-based Models in Real-Time Fusion Plasma Control Systems

Achieving reliable real-time control in fusion plasma experiments requires strict timing guarantees across entire control algorithms. In earlier work by Abbate et al. (2023), we demonstrated the feasibility of neural-network-based control algorithms on the DIII-D tokamak using the internally developed open-source Keras2C library for model conversion into C (Conlin et al. (2021)). However, the initial implementations relied on data buffering and branching logic outside the neural network code, causing variability in execution times. Subsequent deployments on DIII-D and KSTAR—including the RTCAKENN algorithm for kinetic profile reconstruction—proved that minimizing branching and buffering throughout the pipeline yields consistent millisecond-level cycle times under real experimental conditions (Shousha et al. (2023)). However, keeping pace with rapidly evolving AI frameworks (e.g. PyTorch) is challenging. Finally, we, therefore, propose a community-driven open-source effort to expand the tool, enabling real-time deployment across diverse systems that require strictly bounded execution times.. Authors: Shousha, R. [Princeton Plasma Physics Laboratory (PPPL), Princeton, NJ (United States)] (ORCID:0000000314988980); Steiner, P. [Princeton Univ., NJ (United States)]; Seo, J. [Chung-Ang Univ., Seoul (Korea, Republic of)]; Conlin, R. [Univ. of Maryland, College Park, MD (United States)]; Abbate, J. [Princeton Plasma Physics Laboratory (PPPL), Princeton, NJ (United States); Princeton Univ., NJ (United States)]. DOE Contract: AC02-09CH11466; AR0001166; FC02-04ER54698; SC0015480; SC0015878; SC0021275. Subjects: AI-based models; Fusion plasma; Kinetic profile reconstruction; Neural networks; Plasma diagnostics; Real-time control