Model Predictive Control For Permanent Magnet Synchronous Motor Based On Discrete Space Vector Modulation

The model predictive current control(DSVM-MPCC)based on discrete space vector modulation has the advantages of fast dynamic response,easy implementation,and containing nonlinear constraints,etc.It is gradually applied to the control of permanent magnet synchronous motors(PMSM).However,the traditional DSVM-MPCC has the disadvantage of a large amount of calculation when selecting the voltage vector.To solve this problem,a method based on the actual voltage vector preselection is proposed to reduce the amount of calculation when selecting the voltage vector.Secondly,DSVM-MPCC is susceptible to the uncertainty of the motor model parameters.To solve this problem,a model predictive current control based on recursive least squares and discrete space vector modulation(RLS-DSVM-MPCC)is proposed for online model parameter update to improve the accuracy of model prediction.First,a brief overview of PMSM is given.First,the structure and classification of PMSM are introduced,and the similarities and differences between PMSM and BLDCM are compared.Second,the basic working principle of the three-phase inverter is explained,and the commonly used coordinate transformation equation and the relationship between the switching state and the output voltage are derived.Third,a dynamic mathematical model of PMSM and a nonlinear equivalent circuit model based on Ansys are established,and the equivalent circuit and structural block diagram of the model are built.Secondly,a DSVM-MPCC method based on effective voltage vector preselection is proposed to improve the control performance of PMSM and reduce the calculation amount of DSVM-MPCC.First,the principle and realization of space vector modulation(SVM)and field-oriented control(FOC)are introduced.Second,the principle and implementation of discrete space vector modulation(DSVM)and model predictive current control(MPCC)are introduced.The actual voltage vector preselection method is used to firstly determine the optimal area,and then the optimal virtual voltage vector is selected in the optimal area.The 38 candidate voltage vectors are reduced to 13 by this method,which reduces the amount of calculation for voltage vector selection.Third,the simulation models of the two methods are built in Simulink.The simulation results show that the proposed DSVM-MPCC has better control performance than the traditional FOC.Then,the RLS-DSVM-MPCC is used to improve the accuracy of model prediction in order to solve the influence of the motor parameter uncertainty on the DSVM-MPCC.First,the principle of online parameters estimation of PMSM based on RLS is introduced.Second,the RLS method is applied to the DSVM-MPCC,and the switching conditions for the model parameter update are given.Third,a simulation model was built in Simulink to study the static and dynamic parameter estimation accuracy and control performance of the RLS-DSVM-MPCC when the parameters are uncertain.The simulation results show that the proposed RLS-DSVM-MPCC reduces the influence of parameter uncertainty on model prediction and improves the robustness of the control system.Finally,a set of 24 V PMSM hardware experiment platform is built on the basis of theory and simulation.First,the hardware circuit design of PMSM is introduced,including control circuits,drive circuits,detection circuits and step-down circuits.Second,the timing relationship of software execution and the host computer interface are introduced.Third,the FOC is deployed on the hardware experiment platform.The experimental results show that PMSM has small speed error and torque ripple,which verifies that the designed hardware and software are correct and effective.