Machine learning models for real-time, high bandwidth inference of ELM events and confinement regime with 2D BES at DIII-D

Multi-channel fluctuation diagnostics capture the spatial patterns of high-bandwidth plasma dynamics. Here, we report on an effort to develop machine learning (ML) models for the real-time identification of edge-localized-mode (ELM) events and the turbulence properties of confinement regimes using the 2D Beam Emission Spectroscopy (BES) system at DIII-D. The "edge ML" models will be deployed on a high-throughput FPGA accelerator for integration in the real-time plasma control system (PCS). The models will generate reduced signals that correspond to ELM activity and turbulence dynamics, and the real-time PCS will learn to avoid ELM regimes and to steer the plasma towards and maintain advanced confinement regimes such as the wide pedestal QH-mode. The 2D BES system captures plasma density perturbations imprinted in neutral beam emission at a 1 MHz frame rate. The edge ML models will analyze about 10 ms histories from the BES data stream to assess ELM and turbulence activity. Preliminary results for classifying active ELM events give a ROC-AUC score of about 0.98 for validation data. We also explore different neural network architectures such as autoencoders to compress spatio-temporal information in low-dimension feature space for multiple classification and prediction tasks.