Overview of High Accuracy, Multi-device Disruption Event Characterization and Forecasting (DECAF) Research

Physics-based disruption event characterization and forecasting (DECAF) research [1] determines the proximity of tokamak plasma states to a critical disruption warning level providing early forecasts for disruption avoidance or to cue mitigation. Offline analysis accesses the full databases of several tokamaks (e.g. KSTAR, MAST/-U, NSTX/-U, ASDEX-U, DIII-D, ST-40) providing understanding, validation, and extrapolation of models for future devices. Fully automated analysis of large datasets is possible with results showing true positive rates over 99%. Analysis of vertical displacement events shows high prediction accuracy: true positive/negative rates of 61.7% / 38.0% - a combined true accuracy rate of 99.7%. Density limit investigations show that plasmas disrupt after crossing microinstability limits [2] before reaching the Greenwald limit. Real-time (r/t) DECAF experiments on KSTAR produced over 50 plasmas forecast with 100% accuracy in r/t, some triggering controlled plasma shutdown, disruption mitigation, or avoidance actuators. Warnings were issued well before (~1.0s) the expected disruption. R/t magnetics, T_e profiles from electron cyclotron emission (ECE), 2D T_e fluctuation data from ECE imaging, and velocity profile acquisition are installed, with MSE to follow.

[1] S.A. Sabbagh, et al., Phys. Plasmas 30 (2023) 032506; https://doi.org/10.1063/5.0133825

[2] M. Bernet, et al., PPCF 57 (2015) 014038; M. Giacomin, et al., PRL 128 (2022) 185003