Off-normal warning threshold development on SPARC

To achieve a high fusion performance, SPARC tokamak plasmas will have large stored energies, which could lead to high heat fluxes on in-vessel components and large forces applied to conducting structures in the event of a disruption. This work explores the preliminary development of the Off-Normal Warning system for SPARC, the aim of which is to minimize the disruption rate via the detection and pacification of anomalous conditions. The detection will be facilitated via physics-based warning thresholds as well as machine learning-based Proximity-to-Instability Algorithms, and the pacification (e.g. equilibrium steering, "soft-landing" triggers, or DMS triggers) will depend on the severity and (more notably) the type of the anomaly. An emphasis is placed here on the development of physics-based warning thresholds for two of the disruption types expected for the first L-mode campaign on SPARC, radiative instabilities (i.e. UFOs and impurity accumulation) and Vertical Displacement Events. Additionally, the diagnostic requirements for detecting these anomalies based on expected physical timescales for SPARC are presented.