Research On Explicit Model Predictive Control Technology Of Pmsm

Model predictive control(MPC)has better dynamic performan ce than classical PID control,but for a long time,it can not be applied to the fast system because of the computational complexity.The presentation of explicit model predictive control(EMPC)provides a solution to the problem of computational complexit y.Its successful application is of great significance to the improvement of PMSM dynamic performance.This paper mainly focuses on the following aspects of EMPC:Firstly,the PMSM mathematical model is established,and the PMSM vector control strategy is given.The traditional cascaded PI controller is analyzed and the system can not achieve high dynamic response in the case of small inertia system with cascaded structure.In order to improve the dynamic response of the system,a cascade-free structure predictive controller is proposed to control the motor speed,and the PMSM prediction model is derived in detail.Secondly,there are two constraints of the maximum current and maximum voltage in the motor drive control system.The process of MPC and PI handl ing the constraints is compared and analyzed.It is concluded that the limiting method of PI control is not optimal.In order to better solve the constraint problem in MPC,EMPC is used to transform the problem into quadratic programming offline,and the solution process is deduced in detail.According to the system control target and prediction model,the EMPC cost function including variable speed error,q axis current and control increment is given.The pole distribution law of the closed-loop system is obtained,and the EMPC stability analysis is carried out in the basis,three key parameters tuning methods of the controller is also determined.The results of analysis are verified by MATLAB.Then,the EMPC robustness analysis for measurement noise and mo tor parameters is given.The influence of stator resistance deviation,stator inductance deviation and rotation inertia deviation on the dynamic performance and stability margin of the system is analyzed.The law of the influence in different controller parameters is given,and a controller parameter tuning method considering both the rapidity and robustness of the system is proposed.In order to further improve the anti-disturbance capability of the speed control,the predictive control model containing the load torque state is established.The Kalman filter(KF)is used to observe the load torque,and the compensation is made in the prediction of the PMSM speed response in order to improve the dynamic response capability of the speed when the load torque fluctuates.Finally,the hardware platform with XMC4500 as the control core is built,and the realization of EMPC algorithm is achieved by reducing the software size of EMPC algorithm by measures of reducing the complexity.The experimental results show that,compared with the PI control,the PMSM controlled by EMPC has the advantages of faster dynamic response and no overshoot,which shows a better tracking ability for triangular wave and sinusoidal wave,and can recover a given speed faster when the load is added.It is proved that the EMPC design method proposed in this paper can obtain a PMSM system with good dynamic and robust.It provides guidance for the design of EMPC controller of PMSM and extends the application scope of EMPC.