Research On Predictive Torque Control Strategy Of Permanent Magnet Synchronous Motor

Permanent magnet synchronous motor has the advantages of small size,fast response and reliable operation,so it is widely used in industrial and agricultural production,military and daily life.The control method is an important factor to determine the performance of PMSM speed regulation system.However,the traditional control method has been difficult to meet the requirements of high performance control,so it is necessary to conduct in-depth research on the high performance control method of PMSM.Direct torque control(DTC)has the advantages of simple structure,fast response and strong robustness,so it is widely used in the high performance drive system of permanent magnet synchronous motor.The traditional direct torque control(DTC)does not need a modulator but selects the appropriate basic voltage vector according to the error of current torque and flux linkage by looking up the table to control the motor.However,due to the limited number of basic voltage vectors,there is a problem of large torque ripple in the traditional DTC.Predictive torque control is a high-performance control strategy that combines predictive control with direct torque control.This paper studies predictive torque control,and the specific research contents are as follows:Firstly,in view of the problem that the deadbeat control cannot track the actual torque and flux flux due to the PI controller’s inability to output the reference torque quickly and accurately in the case of load mutation or parameter perturbation in the traditional DBDTC method,a DBDTC method based on anti-disturbance sliding mode(ASMC)is proposed in this paper.The ASMC includes an extended sliding mode disturbance observer(ESMDO)which can estimate the system disturbance and compensate it to the sliding mode controller,so as to provide DBDTC with a fast and accurate reference torque and improve the system anti-disturbance ability.In addition,in order to further improve the performance of ASMC-DBDTC,a variable exponential power reaching law is designed,which has a faster convergence speed and the ability to reduction chattering,and the reaching law can be used in the sliding mode controller to obtain better dynamic,steady-state performance.Then,in order to solve the problems of large torque ripple and large amount of calculation in traditional model predictive torque control(MPTC),this paper proposes a deadbeat MPTC method based on discrete space vector modulation(DSVM).This method uses DSVM to synthesize a large number of virtual voltage vectors,and uses deadbeat technology to calculate the reference voltage vector to avoid enumerating all the virtual voltage vectors.Then,three candidate voltage vectors are calculated by performing algebraic operations on the reference voltage vector,so that the method reduces the number of predictions to three while increasing the selectable voltage vector,and reduces the torque ripple and the amount of calculation.In addition,in view of the problem that the weight coefficient of the traditional MPTC cost function are not easy to tune,and the torque and flux linkage need to be predicted online,this paper proposes a cost function without weighting factor based on reference voltage tracking error.The cost function only needs to be substituted into the already calculated reference voltage vector and the duty cycle of the candidate voltage vector,without parameter tuning and online prediction of torque and flux linkage,which further reduces the amount of calculation.