Research On Pmsm Model Predictive Control Strategy With Dynamic Weight

Permanent Magnet Synchronous Motor(PMSM)is used in the field of industrial production,intelligent manufacturing,and aerospace because of its simplified design,high control accuracy,and stable performance.However,with the technological advancement of the society,higher requirements are made for the precision,promptness and consistency of the control of PMSM.For traditional motor control strategies such as magnetic field directional control and direct torque control,it is necessary to keep pace with the development of the times and combine with modern advanced control algorithm to provide a better control efficiency for PMSM.This paper takes a two-level voltage-source inverter-driven PMSM system as the controlled object,and deeply investigates the shortcomings of model predictive control in the field of PMSM control,and research around the difficulties of designing weight coefficients and large torque fluctuations in the model predictive torque control system of PMSM.First,using the traditional model prediction torque control strategy,the influence of the adjustment of the weight coefficients on the torque control performance of the PMSM is investigated,and the direction of performing dynamic adjustment of the weight coefficients is determined.Secondly,the dynamic weighting strategy based on fuzzy control is designed for the problem that the weight coefficients are difficult to design in the traditional model prediction torque control strategy of PMSM.By analyzing the influence of torque and magnetic chain on the PMSM system and the role of weight coefficients in the model prediction control system,a fuzzy controller with torque error and magnetic chain error as fuzzy input quantities and weight coefficients as fuzzy output quantities is established.Finally,the simulation results prove that the dynamic weighting strategy based on fuzzy control can solve the problem that the weight coefficients are difficult to design and realize the online adjustment of the weight coefficients,which improves the dynamic performance of the system torque control although the performance in suppressing the motor torque pulsation is not significantly improved.Then,for the problem of large torque pulsation in the model prediction torque control strategy of PMSM,the dynamic weight model prediction torque control strategy with torque pulsation optimization is designed.In the structure of the traditional model prediction torque control cost function,a new cost function is established by introducing the torque pulsation optimization term,and the torque ripple is designed as the criterion for judging the motor torque control performance,and the relationship between the torque ripple and the weight coefficient in the cost function is established,and the dynamic optimization design of the weight coefficient is carried out based on the torque ripple minimization method.Finally,through simulation,the feasibility of the dynamic weight model prediction torque control strategy based on fuzzy control and torque ripple optimization is verified to improve the control performance of motor torque ripple,and make up for the deficiency that the traditional model prediction torque control strategy based on the cost function cannot achieve the effective suppression of torque ripple by adjusting the weight coefficients.Finally,the design of the PMSM control system is completed,the experimental platform of the PMSM is built,and the experiments on the sudden change of motor load torque under the two dynamic weight prediction torque control strategies of the PMSM model designed in this paper are completed,and the experimental results verify the feasibility of the dynamic weight model prediction torque control strategy based on fuzzy control in effectively suppressing the dynamic performance of the motor torque and the effectiveness of the torque pulsation.The experimental results verify the feasibility of the fuzzy control-based dynamic weight model predictive torque control strategy in effectively suppressing the motor torque dynamic performance and the effectiveness of the optimized dynamic weight model predictive torque control strategy in improving the motor torque pulsation.