An improved method of model predictive current control for multilevel cascaded H-bridge inverters

Abstract

Abstract Finite control set ModelPredictive control (FCS-MPC) with the principle of considering all voltage vectors to find the optimal voltage vector for multilevel inverter in a very small sampling cycle is hardly feasible because there is no modulation part, the implementation of optimizing common-mode voltage and switching number for the multilevel inverter should be performed in the cost function. To solve the above problem, this paper proposes an improved method of model predictive current control selecting 19 adjacent voltage vectors and using weighting coefficients for common-mode voltage elimination and switching optimization. By using a discrete-time model of the system to predict the future value of the current for the voltage vector in the previous sampling cycle and its 18 adjacent voltage vectors, the one that minimizes a cost function will be selected. Thus, in a multilevel inverter with any number of levels, the cost function is performed only 19 times in a sampling cycle. The computation on FPGA allows 19 calculations of the cost function to be performed in parallel, so the executing time is very small. The feasibility of the proposed algorithm is verified by simulation model on MATLAB-simulink software and the experimental 11-level cascaded H-bridge multilevel inverter model.