Edge-Localized Mode Detection and Correlation with Rotating MHD modes for Disruption Event Characterization and Forecasting

Edge-Localized Modes (ELMs) are transient instabilities that eject heat and particles from the edge of a tokamak plasma onto its walls. While typically not directly disruptive, ELMs can trigger more detrimental plasma instabilities that can disrupt plasma confinement. ELM identification is hence an important capability to determine the threat ELMs pose to plasma termination. The Disruption Event Characterization and Forecasting (DECAF) code works to resolve, characterize, and forecast event-chains that lead to disruptions, including disruptive event “seeds”. A newly developed DECAF capability to robustly and reliably identify ELMs using several plasma signals as input is presented. The detection algorithm uses D_⍺ light to find D_⍺ emission transients and electron temperature profiles to critically distinguish edge-localized events from global ones by processing the profile evolution through the mode dynamics. The presented ELM detection capability was validated on a database of ELMing and non-ELMing KSTAR and NSTX plasmas. Further, using the DECAF event chain analysis framework and its existing rotating MHD event module, we apply the novel ELM identification capability to preliminarily study the extent of correlation between ELM-events and rotating MHD-events.