Statistical Performance Verification of FCS-MPC Applied to Three Level Neutral Point Clamped Converter

In this paper we demonstrate how a statistical model checking approach can be used to check the dynamic performance of the finite set model predictive control algorithm for a standalone 3-level neutral point diode clamped converter. The robustness of the control algorithm under parameter uncertainty is also analyzed. Finite control set model predictive control (FCS-MPC) algorithm has found many applications in power electronics due to the straightforward control design and the possibility to include different control objectives. The control algorithm for 3-level neutral point diode clamped (NPC) converter has to address several objectives to provide optimal reference tracking during load transients. Therefore, looking from the perspective of the implementation, the FCS-MPC algorithm suits the control requirements of NPC converter. However, the problem remains in performing an analytical performance verification of the algorithm to demonstrate its robustness, which is compulsory for any industrial application. In this paper, we present how a statistical model checking approach can be used to solve this problem and also provide valuable data about the algorithm’s performance during transients and in the case of parameter uncertainty. A benchmark model is created in Matlab/Simulink to validate the correct system modeling in UPPAAL SMC toolbox.