Adaptive control of rotating magnetic perturbations in HBT-EP using GPU processing

Abstract

Feedback control has become a crucial tool in the research on magnetic confinement of plasmas for achieving controlled nuclear fusion. We present the first experimental results from a novel feedback control system that, for the first time, employs a graphics processing unit (GPU) for microsecond-latency, real-time control computations. The system was tested on the HBT-EP tokamak using an adaptive control algorithm for control of rotating magnetic perturbations. The algorithm assumes that perturbations of known shape are rotating rigidly, but dynamically derives and updates the rotation frequency to improve phase and gain accuracy of the control signals. Experiments were set up to control four rotating n = 1 perturbations at different poloidal angles. The perturbations are treated as coupled in frequency but independent of amplitude and phase, so that the system effectively controls a helical n = 1 perturbation with unknown poloidal spectrum. The control system suppresses the amplitude of the dominant 8 kHz mode by up to 60%. Deviation from the optimal feedback phase combines suppression with a speed up or slow down of the mode rotation frequency. The feedback performance is found to exceed previous results obtained with an FPGA- and Kalman-filter based control system without requiring any tuning of system model parameters.