| The electrification of automobiles can effectively reduce energy consumption and pollution emissions.It is one of the important directions of the new four modernizations of automobiles,and it is of great significance for our country to become a powerful automobile country.With the development of electrification of automobiles,the application of electric motors in automobiles has become more and more extensive.Permanent magnet synchronous motors(PMSM)have the advantages of high power density and high efficiency,and are widely used in the field of automotive motors.The high-performance motor control strategy plays an important role in the PMSM control system.Deadbeat control(DB),as a new type of motor control strategy,can achieve the "zero error" control of the target theoretically,and it is currently a research hotspot in the field of motor control.This paper focuses on the research on the flux and torque deadbeat control strategy of the PMSM for vehicles.The main research work is as follows:The paper simplifies the flux and torque variation model of permanent magnet synchronous motor.Based on the simplified model,the mathematical model of flux and torque deadbeat control is established.The ideal voltage vector expression for deadbeat control is derived,and the space vector modulation(SVM)is adopted to generate the ideal voltage vector.The simulation result verifies the feasibility of flux and torque deadbeat control.When the ideal VV exceeds the space vector modulation(SVM)range,this paper adopts an improved ideal voltage vector generation strategy to reduce the switching frequency of the system.The simulation result verifies that the improved ideal voltage vector generation strategy can effectively reduce the average switching frequency with the stable control performance.The model predictive control is applied to the flux and torque deadbeat control to output the voltage vector,and the cost function representing the error of the voltage vector is designed.The voltage vector that minimizes the cost function is selected from the two basic voltage vectors,and simulation verification is carried out.In order to avoid cost function calculation,a simplified selection method of voltage vector that only judges the amplitude and angle of the ideal voltage vector without cost function calculation is further proposed.Simulations and experiments have verified that the simplified selection method can effectively reduce the calculation time and improve the real-time performance while keeping the control performance basically equivalent.A voltage vector expansion method based on discrete angle and continuous amplitude is proposed.The three-phase switching duty cycle of the expanded voltage vector can be obtained by looking up the table and calculating the expansion ratio of the voltage vector amplitude without space vector modulation.The simulation results show that the deadbeat model predictive control based on the extended voltage vector can effectively reduce the torque ripple of the system.In order to reduce the calculation complexity of ideal voltage vector angle and torque angle,two BP neural networks for calculating ideal voltage vector angle and torque angle are obtained by constructing data sets and network training.The ideal voltage vector angle and torque angle are predicted based on BP neural networks instead of complex mathematical models.The simulation verified the accuracy of the BP neural network angle prediction results and the feasibility of the flux and torque deadbeat control based on the BP neural network angle prediction. |
