Finite set model predictive control of a flying capacitor converter with a geometric computational optimization

The multilevel converter has many advantages, and for that reason, they are widely used in medium and high power applications. On the other hand, finite set model predictive control (FS-MPC) is presented as a suitable control strategy for multilevel converters because it can manage many control goals with a simple cost function. However, FS-MPC needs to evaluate all of the voltage vector's actuation in the power converter, and in multilevel topologies this can result in a high computational effort. This paper proposes a geometrical method based on the predictive model to reduce the number of vectors that must be evaluated when using the FS-MPC strategy. The method is validated using a four-level three-cells flying capacitor converter, that has 512 voltage vectors. In its poorest performance, the proposed method reduces the computational effort to 35% of that used to evaluate the complete case (512 voltage vectors). Simulation results are presented to validate the proposed control strategy.