Research On The Model Predictive Torque Strategy Of Permanent Magnet Synchronous Motor

Permanent magnet synchronous motor(PMSM)has been extensively utilized in many industrial areas due to its high efficiency and strong reliability.In recent years,with the continuous development of power electronics,model predictive control(MPC)has become a highly potential PMSM high-performance control strategy because of its simple structure and strong flexibility.As a classic control strategy in MPC,model predictive torque control(MPTC)has inherited its advantages and still faces many problems that need to be improved.For this reason,this paper takes the PMSM drive system fed by a 2L-VSI as the research topic,and focuses on the finite control set-MPTC(FCS-MPTC)strategy,mainly discussing its large torque ripple and poor parameter robustness and other issues.Also,this paper adopts a one-shot delay compensation method to solve the control delay problem that is widespread in digital systems.Aiming at the problem of severe torque ripple in the conventional MPTC strategy,this paper designs a MPTC without weighting factor based on discrete space vector modulation(DSVM).This strategy uses DSVM technology to form an expanded control set to improve control accuracy.The prediction model is established based on the deadbeat control principle,and the optimization problem is designed and optimized.Then a cost function containing only the error term of the voltage vector is constructed,which avoids the introduction of weight factor.It is verified by simulation that this strategy can significantly improve the control effect of torque and flux.Aiming at the problem of strong parameter dependence in the conventional MPTC strategy,a robust MPTC strategy based on predictive error compensation(PEC)is proposed.The core idea of this strategy is to store the current prediction error by establishing a look-up table(LUT),and design an improved update mechanism to ensure the timeliness of the current prediction error stored value,so as to reasonably compensate for the forecasting process.It is verified by simulation that this strategy has strong robustness to parameter changes.Finally,this paper carries out experimental verification based on the 750W PMSM drive system experimental platform.By comparing the experimental results of dynamic and steadystate performance under different control strategies,the effectiveness of the two improved control strategies are proved.