Abstract

The reliable operation of burning tokamak plasmas will require fast models to accelerate scenario development, to validate controllers, to implement real-time, model-based proximity control strategies and to augment the real-time estimation of the plasma state with the reduced diagnostic suite that is compatible with harsh fusion reactor conditions. In this talk, we present an encompassing simulation workflow based on the RAPTOR transport solver, evolving the plasma profiles dynamics by solving the coupled dynamics of current density diffusion, heat and particle diffusion, and the MEQ free boundary equilibrium suite. This coupled workflow is routinely used at TCV for shot-preparation, real-time control and post-shot interpretation. RAPTOR has been coupled to the inverse equilibrium solver FBT, the forward equilibrium solver FGE and the equilibrium reconstruction code LIUQE, respectively to design more accurate feedforward poloidal field coil traces, to validate magnetic controllers in flight simulator mode and to obtain improved kinetic equilibrium reconstructions. Both in the real-time control system and during post-shot analysis, an Extended Kalman Filter (EKF) is used to assimilate raw Thomson scattering measurements into the RAPTOR simulation. Additionally, the EKF estimates disturbance source profiles and model parameters to account for systematic model-reality mismatches, providing an avenue to automatically validate simple transport models that are easily interpretable and machine-independent. Finally, we will show how the real-time electron temperature estimate, informed by real-time TORBEAM for ECRH sources, has been successfully used for local electron temperature profile control.