Model Predictive Parameter Mismatch Control Of VIENNA Rectifier

Three The three-phase three-level Vienna rectifier has the advantages of high power factor,low stress of power switch,no dead zone and high reliability,so it is favored in aerospace,transportation,new energy vehicles,microgrid and other industrial fields.The finite control set model predictive control(FCS-MPC)is robust,does not need complex parameter tuning,has fast dynamic response,and can optimize multiple control objectives at the same time.However,the performance of FCS-MPC depends heavily on the accuracy of the system model,and the long-term operation parameters of the system will change,resulting in FCS-MPC parameter mismatch.Therefore,it is of great theoretical significance and engineering application value to explore a control method for FCS-MPC parameter mismatch of Vienna rectifier and improve the performance of Vienna rectifier.The simplified model predictive control(SMPC)for Vienna rectifier is studied strategy,in view of the problem that the switching frequency is not fixed and the harmonics are scattered in the traditional model predictive control of Vienna rectifier,a constant frequency model predictive control(CFMPC)method is studied by combining the model predictive control and modulation of Vienna rectifier.In addition,when the system has parameter mismatch,the process of vector selection under SMPC strategy is analyzed,and the effect of parameter mismatch on system performance is discussed.Aiming at the problem of parameter mismatching under the two model predictive control strategies of SMPC andCFMPC,the paper studies the parameter mismatching control methods of inductor observer of SMPC strategy based on Vienna rectifier and Luenberger disturbance observer of CFMPC strategy respectively.Based on the mathematical model of the system,the observer is constructed and applied to the simplified model predictive control current inner loop.The model predictive parameters of the system are corrected in real time by the inductor observer,so as to eliminate the adverse effect of parameter mismatch on the system.The Luenberger disturbance observer is introduced into the constant frequency model predictive control,and the parameter change of the system is regarded as a disturbance.The prediction model of the real-time compensation system for the disturbance change is observed by the Luenberger disturbance observer,which solves the problem of the decline of the system parameter mismatch performanceThe experimental prototype of Vienna rectifier is built,and the two algorithms of SMPC and CFMPC are verified on the prototype.The experiment shows that Vienna rectifier has good steady-state and dynamic performance under the two control strategies.The results show that when the system parameters are mismatched,the inductor can observe the inductance change of the system in real time and correct the prediction model parameters of the system,and the Luenberger disturbance observer can observe the prediction model of the disturbance change and real-time compensation system.Both of the two control methods can effectively suppress the adverse effects of parameter changes on the system.