Research On Model Predictive Control Method For NPC Three-Level Grid-Connected Inverter Based On Optimal Switching Sequence

Model predictive control(MPC)has the advantages of simple principle,fast dynamic response,and can be applied to non-linear systems,and has become a research hotspot in recent years.However,the switching frequency of MPC is not fixed and the control accuracy is low,which makes it difficult to design the parameters of the grid-connected filter and affects the power quality of the grid.This paper mainly researches the problems of unfixed switching frequency and large tracking error of the model predictive control,and proposes a new model predictive control algorithm to solve the problems.In this paper,the prediction model of DC neutral point voltage and AC grid current of NPC three-level grid connected inverter is established,and the cost function including neutral point voltage control is constructed.Then,the influence of calculation and sampling delay on model predictive control is analyzed and a two-step predictive delay compensation method is proposed for the delay.Moreover,the optimization algorithm is simplified by sector division to reduce the online computational burden.To solve the problem that the switching frequency of traditional finite set model predictive control is not fixed,a model predictive control algorithm based on optimal switching sequence is proposed.Firstly,the vector sequence prediction model of inverter is established and the cost function including the increment of the current and neutral point voltage is constructed.Then,the vector dwell time is calculated by solve the problem of the minimum value of the cost function with inequality constraints through Lagrange multiplier method.Finally,based on the symbols of neutral point voltage and grid current,the optimization algorithm is simplified,which reduces online computational burden.In order to verify the effectiveness of the proposed algorithm,an NPC three-level grid connected inverter prototype is built.The experimental results show that,compared with the traditional finite set model predictive control algorithm,the optimized switch sequence model predictive control algorithm proposed in this paper can concentrate the inverter output spectrum,improve the accuracy of the grid current control,and has better dynamic characteristics.