Data-Enabled Finite Control-Set Model Predictive Control for LC-Filtered Voltage Source Inverters

In microgrid applications, LC-filtered voltage source inverters are popular for their low harmonic output and compact design. The high order characteristic and coupling effect of this inverter calls for advanced control scheme, where finite control-set model predictive control (FCS-MPC) emerges as a potential alternative among various solutions. However, the FCS-MPC paradigm needs an accurate math model to predict control variables while traditional modeling methods are vulnerable to system parameter variations. To cope with this challenge, a newly developed data-enabled FCS-MPC approach is put forward in this paper. By leveraging the dynamic linearization technique, it constructs a mathematical model only based on input and output data, eliminating the need for specific circuit parameters. This approach not only retains the advantages of traditional model predictive control but also demonstrates superior performance in handling system parameter changes and disturbances. The robustness of the proposed method in tracking ability and adaptation to parameter mismatches is verified by comparative simulation tests and experiments on a three-level neutral-point-clamped inverter. Meanwhile, the influence of different linearization constants on control performance is investigated, providing strong support for the practical application of this control approach.