Enhanced Model Predictive Current Control Based on Runge–Kutta Approximation for a Voltage Source Inverter

Abstract

Abstract This paper proposes a model predictive current control (MPCC) method with duty cycle control based on the Runge–Kutta approximation compared to the Forward Euler Approximation for grid-connected three-phase inverters with output LCL filter. Hence, results of proposed MPCC methods alongside of the conventional MPCC have been investigated to find the best strategy. First, all 7 possible switching states have been checked by the discrete-time system model based on two approximation methods to select a state that minimizes the cost function. However, at this stage only one voltage vector is chosen during one control period, which cannot decrease the current ripples to a minimum value. Hence, for having sufficient performance, the sampling frequency is necessary to be selected high. Then, the idea of duty cycle optimization has been introduced by using two voltage vectors (a non-zero and a zero voltage vector) during one control period. Therefore, the duration of two voltage vectors have been defined according to the principle of current error minimization. Finally, the effectiveness of the proposed MPCC method based on the Runge–Kutta approximation has been verified by MATLAB/Simulink and experimental results exhibit a better steady-state performance with less sampling frequency as compared to the conventional strategy.