Implementing Data-Driven Models for Real-Time Detection and Control of Alfvén Eigenmodes at DIII-D

We developed and implemented a first of its kind fully data-driven system into the DIII-D plasma control system (PCS) to detect and control Alfvén Eigenmodes (AE) in real-time. Susceptibility to fast ion-induced AE is a challenge in fully non-inductive tokamak operation, which significantly reduces fast-particle confinement and results in degraded fusion gain. Controlling AEs in real-time to improve fast-ion confinement is, hence, important for future Advanced Tokamak fusion reactors. Experiments show that a neural network (NN) achieves 91% true positive rate with less than 10% false positive rate [NF 62 (2), 026007] in detecting 5 types of AE (BAE, EAE, LFM, RSAE, TAE) using high resolution ECE as well as CO2 interferometry data. To estimate the neutron deficit, we trained an NN that accurately outputs the classical neutron rate using similar inputs to NUBEAM. We also designed a preliminary ML-based proportional control that feedback-controls the neutral beam power to achieve desired amplitude of AE modes. The effect of AEs on fast-ion confinement is measured by analysing the gap in classical neutron rate from our NN-based NUBEAM and the measured neutron rate.