A modular architecture for off-normal event models and implementation in MOSAIC for SPARC

The simulation of off-normal events (ONEs) that occur during tokamak operation is required to estimate the impact of these events on tokamak subsystems, as well as to facilitate the development of advanced controllers that aim to predict, avoid, or mitigate the effects of such events. We will present a candidate meta-data schema and framework for describing ONEs and injecting them into tokamak simulators. This will be implemented in a simulation environment agnostic way in order to accommodate multiple stakeholder’s simulators both internal and across machines. Two simulators will initially be targeted: a lightweight python driven simulator used for magnet disruption loading and power supply modeling, and the MOSAIC simulation environment used in support of control design and qualification for the SPARC tokamak. Models of “kick” events leading to vertical displacement events (VDEs), as well as a model for impurity injection events (UFOs) leading to impurity accumulation and disruption will be presented. These examples of event chains leading to disruption will be captured and implemented in simulation to demonstrate the completeness, generality, and efficacy of this approach.