A Simplified Model Prediction Control Method for PV Grid-Connected Inverter

Abstract

To optimize the control methods of PV grid-connected inverter, a simplified Model Prediction Control(MPC) method is proposed. Firstly, the second-order discretization mathematical model of grid-connected inverter control is established in the d-q coordinate system to find out the mathematical relationship between the controlled quantity and the input quantity and the related physical quantities involved in the calculation. Then, receding horizon optimization is performed according to the MPC method to obtain the physical quantity value. Since these physical quantities are approximately constant in a brief time, a batch update method can be adopted. The simulation experiments are carried out in the Simulink platform to verify the steady-state tracking effect of the method, the responses of slope function and step function. The simulation results show that the proposed method can achieve error- free tracking for steady-state function and slope function. The step function response has a fast response time and no overshoot, but there is A brief period of three-phase unbalanced time. Finally, a physical experiment platform is built, the method is solidified into a grid-connected inverter controller to be executed, and step function response verification is performed. The experiment results show that the proposed method can achieve steady- state error-free tracking, the dynamic corresponding speed is fast, and the overshoot is not obvious.