Speaker
Description
The exhaust in a DEMO-class tokamak requires continuous operation in detachment [1]. In highly radiative detached regimes edge-localized modes (ELMs) may be suppressed, reducing reliance on RMP coils for ELM suppression. However, these regimes are close to radiative plasma limits which trigger disruptions that threaten machine-integrity. In stark contrast to conventional reactors, this necessitates a control system (not just for the exhaust) that guarantees operation within safety-critical limits in presence of disturbances. Such guarantees rely on dynamic models of the entire plasma that: 1) capture the response to actuators and disturbances; 2) connect to reactor relevant sensors; 3) describe safety critical limits; 4) scale to DEMO size reactors.
In this contribution we validate the dynamics in multiple physics-based models using a system-identification experiment in the TCV tokamak [2]. This data class has proven its use in controller design, providing guarantees for stability and performance [3]. We compare: 1) TCM a three chamber model simulating 0D reservoirs [4]; 2) DIV1D using reservoirs combined with a 1D scrape-off layer [5]; and 3) SOLEDGE3X-EIRENE as the high-fidelity 2D plasma-edge simulator [6]. We find that the coupling to a core reservoir and a realistic time-scale for neutrals to ionize allows the models to align with the measurements. Similar validations on other devices (using existing data) should be prioritized over predicting behavior on non-existing devices.
To enable model-based control for DEMO-class devices, the challenge for integrated modeling is to shift from interpretations (on isolated domains) to full device time-dependent simulations that mimic the complexity faced when operating DEMO. We should prioritize quantification of errors in solutions with respect to the operational window, disturbances and control-relevant dynamics. Physics-based models should be used to design and demonstrate controllers (that take advantage of real-time data) with operational guarantees jointly for core and exhaust. To be clear, demonstration means running high-performance detached discharges without ELMS or disruptions with the 99% reliability required in DEMO, ARC, STEP, and ITER.
References
[1] A. Loarte et al. Plasma Physics and Controlled Fusion, 2025.
[2] B. Duval et al. Nuclear Fusion, 64(11):112023, 2024.
[3] T. Ravensbergen et al. Nature Communications, 12(1):1–9, 2021.
[4] J. Koenders et al. 41st Benelux Meeting on Systems and Control, 2022.
[5] G. Derks et al. Plasma Physics and Controlled Fusion, 66(5):055004, 2024.
[6] H. Yang et al. Plasma Physics And Controlled Fusion, 65(12):125005, 2023.
| Speaker's title | Mr |
|---|---|
| Speaker's Affiliation | DIFFER, Eindhoven |
| Member State or IGO | Netherlands, Kingdom of the |
