Research On Direct Torque Control Technology For Permanent Magnet Synchronous Motor In Electric Vehicles

Vector control and direct torque control(DTC)are two main control methods of PMSM in electric vehicles.Direct torque control has the advantages of simple structure,fast dynamic response,insensitivity to motor parameters,etc.,but it also has the disadvantages of large torque and flux ripple,unsteady switching frequency and so on.Based on that,in this paper,model prediction control(MPC)and sliding mode control(SMC)methods are applied to the direct torque control technology of permanent magnet synchronous motor.After theoretical analysis and simulation verification,the control method can effectively solve the problems of traditional direct torque control.This paper firstly introduces the structure of the PMSM control system of electric vehicle,and analyzes the mathematical models of PMSM in different coordinate systems.At the same time,the advantages of DTC method compared with Vector Control method in the application of electric vehicle are analyzed,and the existing problems and improvement direction are analyzed.Secondly,aiming at the shortcomings of the DTC method due to the high torque and flux pulsation caused by the low efficiency of the electric vehicle and the serious wear of the mechanical structure,as well as the problem of poor performance of the electric vehicle at low speed,a model predictive direct torque control(MPDTC)method is used to control electric vehicle PMSM in this paper.The method performs 8 rolling predictions on the torque and stator flux linkage according to the motor prediction model in each control cycle and the prediction result is sequentially substituted into the objective function order,and the voltage vector that minimizes the error of the objective function is selected as the optimal voltage vector.The method effectively reduces the torque and flux pulsation and improves the low speed performance of the electric vehicle.In order to reduce the computational complexity of the MPDTC method and simplify the design process of the objective function,this paper presents an improved model predictive direct torque control method for PMSM in electric vehicles.The method uses torque and the flux linkage to predict the reference voltage vector of the next cycle and determine the sector to which it belongs,so that the number of predictions is only 3 times in one cycle,which reduces the amount of calculation.By establishing a new coordinate system and determining the equivalent vector,the calculation of the minimum error of the objective function is transformed into the sector judgment of the equivalent vector,which simplifies the system control,then use the principle of compression transformation to judge the region to which the equivalent vector belongs,and finally determine the current optimal base voltage vector and its corresponding optimal switch sequence.By the analysis of MATLAB simulation,the improved MPDTC algorithm not only ensures the dynamic and static performance similar to the traditional MPDTC algorithm,but also reduces the complexity and calculation of the control system.Then,considering the impact of the speed loop on the MPDTC method,a SMC speed controller is designed by using an improved reaching law at constant.This paper analyzes the design process of the controller in detail and expounds the essential reason for its reducing the chattering phenomenon.According to the theoretical analysis,the controller enhances the robustness of the speed loop compared to the PI controller.Based on that,an electric vehicle PMSM MPDTC control system based on SMC speed controller with improved reaching law at constant is designed.The simulation results show that the control system has excellent speed regulation performance,the system dynamic response is faster,the speed is not over-adjusted,and the system has strong anti-disturbance capability,which further reduces the ripple of the flux linkage and torque.Finally,this paper takes TMS320F2812 as the control core to carry out experiments on some of the control algorithms mentioned in this paper,and the hardware circuits and software parts of the control system are designed respectively,and the no-load regulation experiments of the control system of electric vehicles PMSM is completed when the reference speed is 500 rmp and 1500 rmp.