Preemptive tearing mode suppression using real-time ECH steering machine learning stability predictions on DIII-D

Developing and maintaining safe operating regimes are necessary for tokamak-based commercial fusion power plants while tearing modes pose a threat to steady state operation. Future machines plan to have active tearing mode (TM) control by driving current using electron cyclotron heating (ECH), but plans based on previous experiments have the restrictions of wastefully using ECH for TM suppression or needing to wait until a TM appear before performing control. Our approach uses a machine learning TM predictor based on a survival machine framework to get improved long-term TM prediction that we use for control. With this improved prediction, we developed an ITER-like framework for ECH to minimize power used for TM suppression to free up ECH to be used for other advanced scenario tasks such as core current drive. This control framework was deployed in experiment at DIII-D to test the feasibility for active TM control during an advanced tokamak scenario and whether the scenario can be maintained after a possible TM event. While this control was tested in a DIII-D advanced steady-state scenario, TM suppression and control is critical for all tokamak scenarios and this framework provides the flexibility for future experiments to balance ECH tasks to achieve improved feedback control for other tasks beyond the two tasks explored here.