Model predictive control for reduced structure multilevel converters in compact power conversion units

Recently, a branch of multilevel converters is emerged, in which their ‘reduced structure’ topologies use lower number of devices compared to the available topologies. To get a cost efficient converter, lower number of components as well as high quality waveforms, multilevel converters with a ‘reduced structure’ (MCRS) are suitable for low and compact power conversion units such as airplanes, military equipment, locomotives, and satellites. This paper proposes a model predictive control (MPC) for current regulation and DC-link capacitor voltage balancing for the MCRS, where the case study is a three-phase thirteen level packed U-cell (PUC) multilevel converter. The PUC, as a MCRS, can be formed by manipulating either a conventional capacitor-clamped or the cascaded H-bridge multilevel topologies. A discrete-time model of the system is derived and a predictive model-based control is developed according to this model in order to predict the future behavior of the system for all possible switching states; then, one switching state is singled out for the converter that minimizes the introduced cost function. Performance of the proposed MPC strategy for a three-phase thirteen level PUC is evaluated based on simulations with SIMULINK. While predictive control gets involved in huge calculations, simulation assessments confirm the ability of the proposed control scheme.