Jayson Barr

The termination phase of a shot is an essential part of tokamak operations for all machines present and future. In this phase, the plasma current is decreased as much as possible while attempting to avoid disruptions until confinement is eventually lost. Currently, a large fraction of the disruptions that occur during machine operations occur during this termination phase. For the current generation of machines, disruptions are usually tolerable because they cause little damage.

Abstract This paper reports on advances in the state-of-the-art deep learning disruption prediction models based on the Fusion Recurrent Neural Network (FRNN) originally introduced in a 2019 NATURE publication [https://doi.org/10.1038/s41586-019-1116-4]. In particular, the predictor now features not only the “disruption score,” as an indicator of the probability of an imminent disruption, but also a “sensitivity score” in real time to indicate the underlying reasons for the imminent disruption.