January 10, 2026

Paper on Machine-Learning-Based Plasma Confinement State Identification Accepted in Plasma Physics and Controlled Fusion

Our paper, “Plasma Confinement State Classification via FPP-Relevant Microwave Diagnostics,” has been accepted for publication in Plasma Physics and Controlled Fusion. The authors are Randall Clark, Vacslav Glukhov, Georgy Subbotin, Maxim Nurgaliev, Aleksandr Kachkin, Max E. Austin, and Dmitri Orlov. This work presents a parsimonious and robust machine-learning framework for identifying plasma confinement states using a severely constrained diagnostic set relevant to fusion power plants. Relying exclusively on electron cyclotron emission measurements, the approach demonstrates accurate and reliable L-mode / H-mode classification without the need for in-vessel diagnostics, achieving high performance while remaining compatible with the operational constraints of future fusion power plants. The research highlights the potential of minimalist, ML-driven state identification as a key element of resilient plasma control architectures for next-generation fusion devices.


The accepted manuscript is available online at: https://iopscience.iop.org/article/10.1088/1361-6587/ae363c