Permanent Magnet Synchronous Motor Three-Vector Based Model Predictive Current Control With Disturbance Feedforward Compensation

Model predictive current control has attracted more attention of many researchers due to intuitive idea,fast dynamic response and control objectives flexibility.According to the mathematical model of a permanent magnet synchronous motor,the model predictive current control selects the optimal voltage vector through a cost function in a control cycle to drive the motor driven by an inverter.However,because only one voltage vector is selected in one control cycle,and the mathematical model of the permanent magnet synchronous motor is used to predictive the status of the system at the next period.So,the model predictive current control has large steady-state current ripple and strong parameter dependence.Aiming at the problem of large steady-state current ripple,this paper studies a three-vector model predictive current control strategy.Two effective voltage vectors and a zero vector are selected to synth,esize an optimal voltage vector to the inverter in a control cycle.The vector selection is more accurate.Aiming at the problem of poor parameter robustness,this paper studies the disturbance feedforward compensation technology.The speed loop uses a generalized proportional integral observer to obtain the estimated speed in this paper,and the speed disturbance is observed,then compensates for the given value of the q-axis current to improve the parameter robustness of the system.In this paper,the simulation model of three-vector model predictive current control based on disturbance feedforward compensation is built on MATLAB/Simulink,and the experimental result is performed on an experimental platform of a permanent magnet synchronous motor driven by a two-level inverter.Simulation and experimental results show that when the motor parameters are matched,the three-vector model predictive current control and the three-vector model predictive current control based on disturbance feedforward compensation have approximately the same dynamic and steady-state performance.When the motor parameters mismatch,compared with the three-vector model predictive control,the d-axis and q-axis current ripple in the three-vector model predictive control based on disturbance feedforward compensation can be reduced,which proves the parameter robustness of the algorithm.