Title :
Poloidal magnetic flux profile control in tokamaks via normalized coprime factorization robust control
Author :
Barton, Justin ; Ou, Yongsheng ; Xu, Chao ; Schuster, Eugenio ; Walker, Michael
Author_Institution :
Dept. of Mech. Eng. & Mech., Lehigh Univ., Bethlehem, PA, USA
Abstract :
The potential steady-state operation of a fusion tokamak, with good confinement and a high fusion gain, is related to setting up a suitable current density profile in the device. Experiments at the DIII-D tokamak focus on creating the desired current profile during the plasma current ramp-up and early flat-top phases of the discharge with the aim of maintaining this target profile throughout the subsequent phases of the discharge. The time evolution of the current density profile in a tokamak is related to the time evolution of the poloidal magnetic flux profile, which is modeled in normalized cylindrical coordinates by a partial differential equation referred to as the magnetic diffusion equation. Extremum seeking and nonlinear programming techniques have been employed to find optimal open-loop (feedforward) solutions to the finite time control problem during the ramp-up and early flat-top phases. In order to reject the effects of external disturbances to the system, we propose an optimal H∞ feedback control input that is added to the optimal feedforward control input to regulate the poloidal flux profile around the desired reference trajectories of the system. The combined feedforward + feedback, model-based controller is then tested through simulation.
Keywords :
H∞ control; feedforward; magnetic flux; magnetic variables control; nonlinear programming; partial differential equations; robust control; state feedback; H∞ feedback control; feedforward control; finite time control problem; flattop phase; fusion tokamak; magnetic diffusion equation; nonlinear programming; normalized coprime factorization robust control; partial differential equation; poloidal magnetic flux profile control; steady-state operation; time evolution; Equations; Feedforward neural networks; Mathematical model; Tokamaks; Toroidal magnetic fields; Uncertainty;
Conference_Titel :
Control Applications (CCA), 2011 IEEE International Conference on
Conference_Location :
Denver, CO
Print_ISBN :
978-1-4577-1062-9
Electronic_ISBN :
978-1-4577-1061-2
DOI :
10.1109/CCA.2011.6044432