Model-Free Predictive Control Of PMSM Drive System

With the growing energy and environmental issues,the development of pollution-free,sustainable electric vehicle industry has become the consensus of governments and the automotive industry.The model predictive control has been applied to the control of permanent magnet synchronous motor(PMSM)for electric vehicle because of its fast dynamic response and the advantage of dealing with optimization problems,However,model predictive control has weak robustness against parameter variations.Finite control set model based predictive control(FCS-MPC)is needed to improve the steady-state accuracy of controlled system and time-consuming algorithm optimization and simplified calculation.To this end,this paper lays the foundation of theoretical research and provide technical support for the application of FCS model predictive control in electric vehicle PMSM driving system,the research for the realization of PMSM drive system to improve the control performance,have theoretical research significance and engineering application value.The paper first introduces the principle of model-free control.Combining current predictive control,this paper propose the model-free current predictive control drive system which does not depend on the parameter is presented,then model-free current predictive controller is designed,and FCS model-free current predictive controlled SMPMSM drive system is structured,the feasibility and effectiveness of the proposed model based SMPMSM current predictive control scheme is demonstrated by simulation.On the basis of model-free current predictive control drive system,in order to get rid of sensitive dependence on the system parameters of motor completely,differential algebra is used for the permanent magnet flux online identification.Then,this paper propose the model-free predictive control drive system based on SMPMSM which does not depend on the parameters of motor.The system modeling,system simulation and system experiments demonstrate the advantages of the proposed method for the model free torque control.In order to simplify the finite control set model-free predictive control algorithm,at the same time improve the steady-state control accuracy of finite control set model-free predictive control,this paper combines fast vector selection with dual vector duty ratio control,and the fast dual vector model free torque predictive control is proposed,the SMPMSM drive system based on fast dual vector model free torque predictive control is constructed.The system modeling and simulation results show that the proposed control method does not depend on the characteristics of the SMPMSM parameters,and has the advantages of reducing the torque ripple and reducing the computation time.Finally,Aiming at the finite control set model-free control torque predictive control SMPMSM drive system,this paper do research on flexible set cost function of the MTPA control mode and weak magnetic control model.Then,FCS model-free torque predictive controlled SMPMSM control scheme with wide speed range is supposed.Through the system modeling and simulation,it is proved that this control scheme can realize the high speed and high performance of the SMPMSM drive system,and has the advantage of strong robustness.