Torque Ripple Suppression Of Permanent Magnet Synchronous Motor Based On Model Predictive Control

Permanent magnet synchronous motors(PMSM)have gradually become the core of AC speed-regulation system due to its high reliability,large power density,little space occupation and other advantages,and PMSM are widely used in electric vehicles,CNC servo,rail transit and other fields.The predictive control strategy of PMSM has the advantages of fast dynamic respo nse,but due to the influence of factors such as the structure of the motor,the nonlinearity of the inverter and the control error will cause torque ripple in the motor s.Torque ripple can lead to noise and mechanical vibration in the motors and even makes the systems unsuitable for some special occasions.Therefore,the research on torque ripple suppression technology of PMSM predictive control system is carried out.Aiming at the problems that the model predictive control system of PMSM has large steady-state torque ripple under the conditions of inaccurate motor parameters and low control frequency,this paper proposes corresponding improvement strategies for model predictive current control.The main contents of this paper are as follows:Firstly,a simulation model for the predictive current control of a PMSM is established in MATLAB.The motor torque control performance is simulated and analyzed with the presence of inaccurate motor parameters;it has been found that the inaccuracy of the motor dq axis inductance parameters may lead to the steady state torque ripple of the motor.In addition,it is also verified that the model predictive control has a large torque ripple under a lower frequency of the simulation control.Secondly,aiming at the problem that the motor’s steady-state torque ripple increases due to inaccurate motor parameters,a predictive current control strategy for PMSM based on disturbance compensation is proposed.A disturbance observer based on second-order sliding mode is studied;by compensating the disturbance to improve the prediction accuracy,and the torque ripple caused by uncertain motor parameters is reduced.Then,aiming at the problem that the steady-state torque ripple of the PMSMs based on predictive control is large at a low sampling frequency,a new model predictive current control method based on duty-cycle modulation is proposed.The strategy combines the current dead-beat principle with the mathematical model of the PMSM.The three-phase voltage vector is used to divid e the phase plane into three sectors,so the prediction iteration for current is performed three times one cycle under the three sectors,and current prediction iteration is performed to directly obtain the duty-cycle of the three-phase switch drive pulse.The amplitude and phase angle of the voltage vector can be adjusted which can significantly improves the steady-state performance of the motor control system and reduces the torque ripple.Finally,the proposed predictive control method based on the duty-cycle modulation is experimentally verified on the experimental platform of the PMSM prototype.It has a certain suppression effect on torque ripple by comparing with the traditional method under the same condition.