Research On Model Predictive Control Of Three-phase PWM Inverter

With the depletion of traditional energy,exploring and making good use of new energy has become an important trend of human development in the future.Solar,wind and other green renewable energy has gradually become the focus of people’s attention.As a link between new energy and power grid,grid-connected inverter’s control strategy is directly related to the quality of power transmission.Therefore,many scholars root in this,hoping to find an efficient and reliable control scheme to improve the grid-connected power quality of inverter.Among the many control methods,model predictive control(MPC)is favored by scholars because of its advantages of simple and flexible control ideas,fast dynamic response,many controllable variables,and easy to solve nonlinear constraints.A large amount of literature also shows that MPC has excellent control performance in many fields.Therefore,this paper studies the three-phase PWM grid-connected inverter as its control object.This paper first introduces the development background and research status of grid-connected inverter control strategy at home and abroad.Then through the analysis of the basic principle of grid-connected inverter and the derivation of the formula,the mathematical model of grid-connected inverter in three kinds of coordinate system is established,and the principle and realization method of SVPWM modulation mode are expounded.On this basis,the traditional single vector model predictive control and dual vector MPC are used to implement the two control algorithms.Compared with single vector control,dual vector model predictive control overcomes the disadvantages of large current fluctuation,fixed voltage vector direction,less optimization times and poor steady-state performance,and has good control effect.However,this control only controls the alternating current,the current pulse of the straight axis is still large.Aiming at the limitation of this control,based on the dual vector control algorithm,this paper proposes a three-vector model predictive control,that is,a control period is divided into two non-zero voltage vectors and one zero voltage vector to control the grid-connected inverter.The three vector model predictive control uses the across-axis deadbeat method to calculate the action time of the three voltage vectors.Finally,in order to further verify the effectiveness and feasibility of the project,the traditional single-vector model predictive control,two-vector MPC and the proposedthree-vector model predictive control were tested and simulated on the MATLAB/Simulink simulation platform and rt-lab semi-physical simulation platform respectively.The results show that the three-vector model predictive control is more efficient,the current pulsation is smaller,the harmonic content is less,the robustness is stronger,the dynamic performance and steady-state performance of the system are better,it has a good application prospect,the application of multi-vector model predictive control in grid-connected inverter provides certain reference value.