PMSM Model Prediction And Speed Sensorless Control For Electric Vehicles

Facing the increasingly serious environmental problems,there is no doubt that the rise of electric vehicles,and in the electric vehicle industry,motor as the core of the car,has become an eternal topic.Permanent magnet synchronous motor(PMSM),as a kind of high-performance motor,has become a research hotspot of scholars at home and abroad due to its advantages of large torque,high efficiency,simple and reliable structure.In this paper,the control algorithm of PMSM and the technology without speed sensor are studied and improved.It mainly includes the following contents:Firstly,the traditional model predictive current control(MPCC)principle is studied.The traditional MPCC selects an optimal voltage vector through the principle of minimum value function in each sampling period.However,due to the limitation of control set,the traditional single vector model predictive current control has poor stability performance and large current and torque fluctuations.Although duty cycle control makes the effective voltage vector size adjustable in a selection period,there are only six directions,which can not be controlled in all directions.The dynamic and steady performance is still poor.Therefore,this paper proposes an improved method in the dual vector model predictive control strategy.The method selects two voltages in each sampling period.These two voltages can be adjacent complete voltage vectors or adjacent duty cycle voltage.By adjusting the control time,the optimal voltage vector can be controlled,The direction can also be controlled so that the voltage selection coverage is larger and the selection time is shorter.The control principle of q-axis deadbeat is adopted to reduce the torque and current ripple of the control system,and improve the stability of the control system under dynamic and static conditions.The simulation and experimental results also verify the correctness and rationality of the improved MPCC control algorithm.Secondly,based on the sliding mode speed sensorless algorithm of model current predictive control,an improved high-order sliding mode observer is proposed.The observer is improved on the traditional high-order sliding modeobserver,taking the extended back EMF as the state variable,and the improved orthogonal phase-locked loop is used to track the rotor position information,which can improve the stability and observation of the system It can track the change of rotor position and speed quickly.The simulation results show that the improved model can predict the speed and rotor position information more accurately than the traditional sliding mode speed sensor system.The estimation error is small,and the dynamic and steady performance is more prominent.Finally,based on the DSP integrated development platform,the improved MPCC control strategy is verified by experiments.Through the analysis and comparison of the experimental results of the motor under three control strategies,the correctness and rationality of the improved control strategy is verified.