Current Predictive Control Of Permanent Magnet Synchronous Traction Motor At Low Carrier Ratio

In recent years,permanent magnet synchronous motors have gradually replaced asynchronous AC motors as the mainstream motor type used in rail transportation.Traction motors used in rail transit are very powerful,and the switching devices need to work at a lower switching frequency to reduce switching losses;and because the speed of high-speed rail can be 300 kilometers per hour,its traction motors need to work stably at higher speeds.Therefore,considering the working conditions of high speed and low switching frequency,the design of the speed regulation system of the traction motor at the low carrier ratio needs to consider many aspects.This paper analyze the performance of the current-loop of PMSM speed control system at low carrier ratio,then use a new method to improve the performance of current-loop based on current PWM predictive control strategy.Firstly,this paper takes SPMSM as the control object,introduces its working principle and mathematical model,and use engineering methods to calculate the parameters of the speed-loop PI controller and current loop PI controllers of the double closed-loop PI speed control system.The contrast simulation of the PMSM speed control system with high carrier ratio and low carrier ratio is carried out,and then get the conclusion that the double closed loop PI control strategy is not suitable for the PMSM control system at low carrier ratio.Secondly,this paper discusses the acquisition of current prediction model in current PWM predictive control(PPC)system.First,the principle of the traditional current PPC system is introduced.Through analysis,it is found that the delay of the inverter at low carrier ratio cannot be ignored.Therefore,the inverter and the PMSM are integrated as one control subject,then based on the inverter-PMSM integrated model,an integrated current PPC strategy is suggested.Considering the discretization error of the model,the discretization deviation is corrected by estimating the current at 1/2 time of the control period,and an integrated predictor-corrector current model is proposed.Through simulation and analysis,the accuracy,steady-state performance and dynamic performance of the three current prediction models are compared.Thirdly,the problem of model prediction deviation is discussed from two aspects: motor parameter error and voltage command angle error.For the current steady-state error caused by inaccurate motor model parameters,this paper proposes an improved current error integral compensation method by analyzing the relationship between current error and voltage command error in the integrated predictor-corrector current prediction model;for the current steady-state error caused by the mid-voltage command angle error in vector control,this paper proposes an angle compensation strategy suitable for low carrier ratio.Finally,its effect is proved through simulation and analysis.Finally,a PMSM experimental platform based on NI c RIO-9033 was set up to compare the performance of the conventional PPC,integrated PPC and integrated predictor-corrector PPC at low carrier ratio,and the performance of the conventional integral compensation strategy and improved integral compensation strategy was compared in the same situation.