Optimized Finite Control Set Model Predictive Control for a Three-Phase Five-Level Cascaded H-Bridge Multilevel Inverter fed Interior Permanent Magnet Synchronous Machine With On-Line Candidate Switching State Selection

Abstract

Model predictive control is a novel control strategy that is attracting the scientific community due to the several advantages it offers, such as the ability to consider system nonlinearities, the possibility to synthesize a control for a MIMO system instead of multiple SISO, and so on. Control feasibility, due to the very high computational cost required to solve the optimal control problem, is a challenge. By considering electric drives fed by multilevel inverters, the control design is more challenging due to the increased number of available output voltage vectors. In this work, a simple algorithm for the voltage candidate reduction is presented: it allows for reducing the control computational cost, minimizing the switching losses, and minimizing dv/dt on phase voltage waveforms.