Torque Control And MRAS Sensorless Control Based On Dual Model Prediction Of PMSM

Permanent magnet synchronous motors(PMSM)are widely used in high-end manufacturing due to their small size,high efficiency,and high power density.Because traditional vector control is difficult to meet the higher performance control requirements,improved model predictive torque control combined with space vector modulation(MPTC-SVM)is used to improve control performance;in order to further expand the application range of PMSM,model predictive control is also used to improve the control performance of sensorless control based on model reference adaptive system(MP-MRAS),and the effectiveness of the two improvement measures is verified through simulation and experiment.The main content of this article:(1)Starting from the structure and working principle of PMSM,the mathematical model of PMSM is derived,and the model simplification and decoupling are realized through coordinate transformation.The realization of space vector pulse width modulation(SVPWM)is introduced in detail,combined with the mathematical model of the motor to realize the vector control with id=0,and the simulation is carried out in the MATLAB/Simulink environment to verify the feasibility of the traditional vector control(FOC).Lay the foundation for follow-up work.(2)The realization of finite set model predictive torque control(FCS-MPTC)and optimization measures such as one-beat delay compensation and maximum torque-to-current ratio control are introduced.For FCS-MPTC’s complex weight factor setting and larger torque ripple,an improved three-vector model predictive torque control(MPTC-SVM)is proposed:the deadbeat control is combined with the predictive model to calculate the optimal voltage vector and output by SVPWM.Finally,the control effect of MPTC-SVM is verified by comparing FOC and FCS-MPTC through simulation.(3)Introduced the model reference adaptive system and Popov superstability principle,and deduced principle as a PMSM position observer.Aiming at the problem of PI adaptive rate for traditional MRAS,the finite set model predictive control is used as the adaptive rate to realize the model predictive model reference adaptive(MP-MRAS)position observer.The extended finite set by estimated rotational speed is used to remove the restriction of the fixed finite set to the estimated rotational speed,and a piecewise low-pass filter is used to realize rotational speed estimation.Finally,the position observation effect of traditional MRAS and MP-MRAS is compared through simulation,which verifies the improvement of MP-MRAS’s estimation effect.(4)Based on theoretical research,the test verification was carried out on the AC motor drive control platform,and the joint debugging of software and hardware was completed.Combine FOC and MPTC-SVM with traditional MRAS and MP-MRAS position observers respectively,and compare them with experiments.The experimental results verify the control performance of MPTC-SVM and MP-MRAS position observers.