Research On PMSM Torque Ripple Suppression Based On MPC

Permanent Magnet Synchronous Motors(PMSM)are widely used in the field of new energy electric vehicles due to their advantages such as simple structure,small size,and high efficiency.However,factors such as nonlinearity of the PMSM drive system,load disturbances,and sensor errors can cause torque ripple during motor operation,leading to noise and reduced motor lifespan.The Model Predictive Control(MPC)algorithm can fully handle system nonlinearity and disturbances in the process of system modeling and value function,thereby effectively suppressing torque ripple in the motor.The main research content of this paper is as follows:(1)The mathematical model of the PMSM is established,and the two-level threephase inverter and SVPWM modulation technology are analyzed and derived.(2)Conventional model predictive torque control(Conventional-MPTC)and duty cycles based model predictive torque control(Duty-MPTC)are studied.ConventionalMPTC can only select 8 basic voltage vectors,and the phase difference between adjacent basic voltage vectors is 60°,which results in a large deviation between the expected voltage vector and the actual voltage vector that can act on the motor,leading to the problem of large motor torque ripple.Meanwhile,Conventional-MPTC has a large computational load.Duty-MPTC includes a duty cycle modulation module that can change the amplitude of the voltage vector acting on the motor in real-time by the joint action of effective voltage vectors and zero voltage vectors,achieving the effect of torque ripple suppression.However,Duty-MPTC only changes the amplitude of the actual output voltage vector,not the phase of the voltage vector.In addition,the duty cycle modulation module increases the computational load.(3)Section filtrated of extended voltage vectors based model predictive torque control(SFEVV-MPTC)is proposed,which is based on interval selection.SFEVVMPTC determines the space vector interval where the expected voltage vector lies by calculating the deviation between the expected and actual torque of the motor.Within this interval,voltage vector extension is performed based on a fixed voltage vector amplitude and a 15° phase angle.SFEVV-MPTC expands the number of optional voltage vectors from 8 to 26,and then uses interval selection to reduce the number of voltage vectors from 26 to 5 during prediction and calculation.Furthermore,because SFEVV-MPTC adopts fixed voltage vector amplitude and phase angle for extension,it can be designed as a voltage vector table for offline calculation without increasing the additional online computational workload.(4)The effectiveness of the proposed method is verified through modeling,simulation,and preliminary testing on an experimental platform.A simulation model was built and simulation experiments were conducted in the MATLAB/Simulink environment.The simulation results show that,in terms of current harmonics,the current harmonic of SFEVV-MPTC is 8.94%,while those of Conventional-MPTC and Duty-MPTC are 19.51% and 15.84%,respectively.Compared to Conventional-MPTC and Duty-MPTC,the current harmonic content of SFEVV-MPTC decreases by 10.57%and 6.9%,respectively.In terms of torque ripple,the standard deviation of torque ripple of SFEVV-MPTC is 0.1339,while those of Conventional-MPTC and DutyMPTC are 0.3375 and 0.1654,respectively.A hardware testing platform based on TI’s TMS320F28379 D and Ga N power device LMG5200 was built and experimental verification was conducted.The experimental results show that,when the PMSM operates under load in three speed ranges of 300r/min,600r/min,and 1200r/min,the current harmonics of Conventional-MPTC are 49.14%,61.85%,and 55.24%,respectively,while those of SFEVV-MPTC are 32.48%,38.33%,and 35.48%,respectively.Compared to Conventional-MPTC,the current harmonics of SFEVVMPTC decrease by 16.66%,23.52%,and 19.76%,respectively.In terms of torque ripple,the standard deviation of torque ripple of SFEVV-MPTC is 0.0553,0.0548,and0.0545,respectively,while those of Conventional-MPTC are 0.1286,0.1206,and0.1264,respectively.The results of modeling,simulation,and hardware platform experiments demonstrate that the proposed torque ripple suppression strategy is effective.