Application Research For Low Frequency Oscillation Suppression Of Vehicle-grid System Based On Model Predictive Control And Its Improved Algorithm

As power electronic devices are widely used in traction power supply system,the structure of system gradually becomes complicated.The low frequency oscillation(LFO)of the vehicle-grid coupling system occurs from time to time,which has a serious impact on the stable operation of the electric multiple units(EMUs)or locomotivies.At present,the methods for suppressing the LFO of the vehicle-grid coupling system are mainly divided into two categories: the traction network side and the locomotive side,but the former is more difficult to implement.Therefore,this paper chooses optimizing the control strategy of the pulse rectifier to suppress LFO.Combining model predictive control theory and considering the influence of delay,a model predictive control(MPC)based on delay compensation is designed.Subsequently,considering circuit parameter mismatch and system uncertainty,a model predictive control based on deviation correction(DEC)is designed to further improve the decoupling ability and robustness of the system.The proposed methods are compared and verified on the MATLAB/Simulink simulation platform and the hardware in the loop(HIL)experimental platform.The main work of this paper is as follows:Firstly,based on the principle of wire reduction and merging,a reduced-order model of the traction network is built with reference to the transmission line parameters.According to the structure of the CRH5 EMU system,the mathematical model of the dual pulse rectifier in the dq rotating coordinate system is derived.Then,the time delay causes the input admittance of the system to change,which in turn affects the stability of the system.According to the mathematical model of the dual pulse rectifier in the dq rotating coordinate system,a discrete current prediction model is derived,and a two-step predictive control strategy is used to compensate for the delay effect.The cost function is constructed according to the control objective,and the optimal control voltage is obtained through optimization methods.After coordinate transformation and pulse width modulation(PWM),the best switching sequence is obtained.The proposed control strategy is compared and verified on the simulation platform and the hardware-in-the-loop experiment platform.In order to solve the problem that the decoupling effect of MPC is poor and the control accuracy is obviously decreased when the actual system and the model are seriously mismatched,the deviation correction is combined to further improve the robustness of the system.Firstly,the predictive errors caused by the circuit parameter mismatch,sampling and discrete uncertainties are analyzed,and then the deviation correction is constructed according to the predictive error expression.The estimated total system error can compensate the predicted current in real time.Finally,the proposed control strategy is compared and verified on the simulation platform and the hardware-in-the-loop experiment platform.The research in this paper shows that the model predictive control based on delay compensation can effectively make up for the defect of reduced system stability caused by delay.In addition,model predictive control based on deviation correction can improve the decoupling effect of predictive control and improve the robustness of the system.This paper provides two effective methods for improving the model predictive control strategy and suppressing the low-frequency oscillation of the system.