Model-based control development for KSTAR enabled by TRANSP
M.D. Boyer, X. Yuan, F. Poli, H-S. Kim, S-H. Hahn, E. Schuster, S. Morosohk, S. Sabbagh, J-H. Ahn, W.P. Wehner
Division of Plasma Physics (DPP) Annual Meeting of the American Physical Society (APS)
Fort Lauderdale, FL, USA, October 21-25, 2019
Abstract
|
|
Model-based control and scenario development will be critical for
reaching the highest possible performance of KSTAR and future fusion
devices like ITER. Key to the model-based approach is the development
of a hierarchy of models of varying fidelity, execution speed, and
complexity, suited to different roles in the design process. These
approaches are being actively developed for KSTAR to enable reaching
many of the scientific goals of the device, including ramp-up
optimization, non-inductive scenario optimization, and rotation and
current profile control. For this work, both analytic and data-driven
reduction methods have been used, with the TRANSP integrated modeling
code serving as the foundation of the modeling hierarchy. Initial
control algorithms and design tools have been developed that will enable
accelerated offline and real-time model-based decision making in future
experimental campaigns. The reduced models and TRANSP have been
integrated with the control design software Simulink. This enables a
hierarchy of models to be used to validate control algorithms prior to
experiments, which will accelerate the commissioning process. Several
recent results in these areas will be covered along with plans for
future development and applications.