Research On Deadbeat Predictive With Position Sensorless Current Control Strategy For Permanent Magnet Synchronous Motor

Permanent magnet synchronous motor with high efficiency,high ratio of power to weight,reliable operation and other significant advantages,has been widely used and continuously developed in the field of aerospace,Marine electric propulsion,high-end CNC machine tools,industrial robots,electric vehicles and other fields.However,there are still many gaps between China and some developed countries in terms of high-performance inverter products.Therefore,it is of great significance to study the motor control system with high dynamic performance,high precision and strong robustness.The main research work of this paper is as follows:Due to the traditional proportional integral control current loop structure,it’s structure is simple and easy to program,but is not suitable for high-precision control because of its low-pass filtering characteristics,the output of the phase lag,dynamic performance is limited and other shortcomings.Based on the mathematical model of permanent magnet synchronous motor(PMSM)considering delay,the traditional robust current predictive control strategy is analyzed and its performance is analyzed.In order to improve the steady-state current error of this method,a discrete extended state observer error feedback matrix is designed to improve the performance.Then,the proposed algorithm is analyzed from the perspectives of dynamics,stability and steady-state error in the discrete domain.Aiming at the existing method of the deadbeat current prediction,the installation of mechanical position sensor will lead to the problems of large size and weight of the system,easy to damage and high maintenance cost.This paper designed a kind of sensorless control of permanent magnet motor with deadbeat current prediction method,based on model reference adaptive control theory,by constructing feedback gain,there can be deadbeat current forecast and sensorless control together,united into a form of observer,not only can improve the dynamic response of current,and can satisfy the sensorless motor under normal operation.Then,the stability of the proposed method is verified theoretically.According to the transfer function,the closed-loop pole-zero distribution law and trajectory diagram of the observer are analyzed,and the adaptive rate parameters and the feedback gain matrix parameters are designed.The actual permanent magnet synchronous motor(PMSM)is a multivariable and nonlinear controlled object with strong coupling.When the parameters(resistance,inductance,flux,etc.)change,it will have a non-negligible impact on the accuracy and stability of current prediction.To further enhance the robustness of the whole system,in this paper,the proposed sensorless permanent magnet synchronous motor with deadbeat current predictive control method,has carried on the stator resistance,inductance,the sensitivity analysis of parameters,such as rotor flux,through theoretical analysis,the estimated speed error is deduced by the mathematical expression of motor parameter variation,in view of the actual rotor flux affected most of the factors,An algorithm based on Kalman filter was used to identify the rotor flux on line,and the controller parameters were adjusted according to the real-time parameters obtained.In order to verify the proposed current prediction method,the control method of sensenseless permanent magnet motor deadbeat current prediction and the rotor flux identification method online,a 4.4k W permanent magnet synchronous motor towing platform based on STM32F103 VC was built to carry out experiments of on-load acceleration,parameter mismatch and dynamic performance.The results show that the proposed method can effectively reduce the dependence on motor parameters,improve the accuracy of current prediction,enhance the dynamic performance of the current loop,and improve the stability region of the estimated speed and position angle information and the observer convergence performance.