Research On Model Predictive Torque Control And Sensorless Control Of PMSM For Electric Vehicle

Affected by environmental pollution and energy crisis,electric vehicles have become the main trend of automobile manufacturing industry in the future.In electric vehicles,the driving comfort and stability of electric vehicles are determined by the control system of driving motor.Permanent magnet synchronous motor(PMSM)has become the preferred driving motor for electric vehicles because of its high efficiency,high power density and excellent control performance.In this paper,the control algorithm and speed sensorless technology of permanent magnet synchronous motor for electric vehicle are studied and improved,aiming at the motor drive system of electric vehicle.It mainly includes the following contents:Firstly,the direct torque control strategy based on SVPWM is studied.SVPWM can compensate for the errors in the given and calculated values of flux linkage in traditional direct torque control.SVPWM-DTC can reduce the torque and flux ripple of traditional DTC to a certain extent.However,the simulation results show that the torque and flux ripple of SVPWM-DTC is still relatively large.And the dynamic performance and steady-state performance are relatively poor.Secondly,the control strategies of single vector MPTC and two-vector MPTC are studied,and the improvement measures are put forward on the basis of two-vector MPTC control.When determining the action time of two voltage vectors,the torque deadbeat principle is adopted to reduce the ripple of the control system's torque and flux linkage.At the same time,the dynamic and steady-state performance of the system is improved.The simulation and experimental results verify the correctness and feasibility of the improved two-vector MPTC control algorithm based on the principle of torque deadbeat.Then,a simplified MRAS speed sensorless algorithm based on model predictive control is proposed.It is not necessary to locate the rotor position at the next sampling time in the 360 degree range but only necessary to select the eight angle ranges near the rotor position at the previous time.This methodsimplifies the algorithm and improves the operation speed.Through simulation analysis,the improved model predictive MRAS speed sensorless control system has more accurate speed and rotor position estimation results than the traditional MRAS speed sensorless control system.This improved method has less estimation error and excellent dynamic and steady-state performance.Finally,the improved two-vector MPTC control strategy is validated experimentally based on the integrated experimental development platform of DSP.By analyzing the experimental results of the motor under acceleration,deceleration,load-adding and load-reducing conditions,we can see that the motor has good dynamic and steady-state performance in various conditions,which verifies the feasibility of the algorithm.