Disruption Prevention via Interpretable Data-Driven Algorithms On DIII-D and EAST

The Disruption Prediction via Random Forest (DPRF) algorithm is currently implemented in both DIII-D and EAST PCS, and provides not only predictions of impending disruptions in real-time, but simultaneously identifies the drivers of the disruptivity, i.e. feature contributions – all in about 150 - 250 microseconds. This is the first demonstration that a machine learning-based algorithm can provide interpretable predictions in real-time on multiple devices. Providing indications of the ongoing disruption dynamics to the Plasma Control System (PCS) proves to be essential in actuating the proper response to avoid performance loss and disruptions deleterious consequences. On DIII-D, DPRF was upgraded including real-time calculations of profile-based indicators of temperature, density and radiation. Such peaking factors prove to be relevant metrics in impurity accumulation events leading to disruptions in scenarios close to ITER baseline, providing a warning more than 1s prior to disruption. On EAST, DPRF was trained on high density disruptions, and during closed-loop experiments, it has shown to be capable of predicting such cases, and trigger the mitigation system with relevant accuracy. Data-driven models will be integral part of device protection system for ITER and the next generation of devices, but only if capable to demonstrate optimal predictive capabilities and the possibility to reconcile the predictions with the underlying physics dynamics. Current results represent a step forward in data-driven control for scenario optimization and disruption avoidance for ITER and next generation devices. This work establishes the importance of developing tools capable of identifying and informing in real-time the PCS on the dangerous plasma parameters deviations to the disruptive space to enable the proper actuators’ response.