Research On Robust And Fixed Frequency Model Predictive Control Of Grid-connected Inverter

The grid-connected inverter is the power electronic interface between renewable energy generation and the power grid,and its control performance determines the dynamic and static performance and power quality of the grid-connected current.Finite control set model predictive control(FCS-MPC)is a new type of control method that can directly utilize the discrete characteristics of grid-connected inverters without modulation,with fast dynamic response and multi-objective control advantages.And it has been rapidly developed in the field of grid-connected inverter control.In the field of FCS-MPC control of three-phase grid-connected inverters,Simplified Finite Control Set Model Predictive Current Control(S-FCS-MPCC)based on the optimal output voltage has the advantages of a small amount of calculation and simple control logic.Therefore,the thesis takes the grid-connected inverter using S-FCS-MPCC as the research object,and focus on the aspects of enhancing the parameters robustness and steady-state performance of SFCS-MPCC.The main contents and innovations of this thesis are:(1)The research status of FCS-MPC in power electronic converters is introduced.The basic principles of grid-connected inverters FCS-MPCC and S-FCS-MPCC are described respectively.And it is compared with traditional linear PI control and traditional FCSMPCC in the aspects of the the dynamic and steady state performance,parameter robustness,and current spectrum characteristics.(2)Aiming at the problem of poor parameter robustness of S-FCS-MPCC,the influence of the mismatched inductance and resistance parameters on the controller to select the optimal output voltage and the performance of grid-connected current are analyzed.Based on the two-step voltage error equation,an inductance identification method based on optimal output voltage selection and a resistance identification expression based on adaptive weight are constructed,which simplifies the problem of singularity judgment processing in parameter identification and improves the parameter robustness of gridconnected inverter FCS-MPCC.(3)Aiming at the problem that the S-FCS-MPCC switching frequency is not fixed,which leads to the dispersion of current harmonic spectrum and poor steady-state control performance,the fixed-frequency method based on duty cycle optimization is improved,and a fixed frequency control method based on two-degree of freedom voltage vector is proposed.The proposed method extends the freedom of control of the combined voltage vector by increasing the combination of output voltage vectors,and the optimal function time of the dual output voltage vector is obtained by the value function optimization method,which can change the direction and amplitude of the output voltage vector at the same time,and the accuracy of current steady state control is improved.(4)A three-phase grid-connected inverter simulation model based on S-FCS-MPCC is built under MATLAB/Simulink environment,and the effectiveness of the parameter identification methods based on the two-step voltage error equation and the fixedfrequency control scheme based on the two-freedom voltage vector are verified.A gridconnected inverter experiment platform based on the Star Sim real-time simulation system is established,which using TI’s DSP TMS320F28335 as the control core.The S-FCSMPCC algorithm of the three-phase inverter and the inductance and resistance real-time identification algorithms under multiple operating conditions are realized.The feasibility of the proposed parameter identification schemes are verified.