Model Predictive Control for Tokamak Central Solenoid Power Supply

Power Supply (PS) systems for nuclear fusion coils are particularly challenging because of the very high current amplitude (tens of kiloamperes), quite unusual in industry applications. Moreover, they need to guarantee a proper plasma initiation and its magnetic confinement for tens of years. To face such demanding requirements, with focus on the Central Solenoid PS of the Divertor Tokamak Test (DTT) facility, Voltage Source Converters are currently under investigation. In particular, this paper analyzes a parallel-connected H-Bridges DC-DC converter. The proposed control strategy consists in a Sphere-Decoding Algorithm (SDA) Model Predictive Control (MPC). This choice aims at exploiting the very fast transient response of MPC and the computational burden reduction of SDA. In order to implement the SDA-MPC, the mathematical model of the presented PS is firstly introduced and validated. Afterwards, the SDA-MPC simulation results are shown and its performance is compared with an interleaved control strategy.