Sensorless Model Predictive Control Strategy For Permanent Magnet Synchronous Motor At Low Speed

AC Drive system is widely used in various fields.PMSM(Permanent magnet synchronous motor)has been widely used in many fields because of its advanta ges of high efficiency,high power factor,high power density and wide speed range.PMSM control system mostly uses double closed-loop vector control system,including speed outer loop and current inner loop.The control performance of the system mostly depends on the control performance of current loop and accurate rotor position information.At present,most commonly used PMSM control systems have good performance at medium and high speed,but permanent magnet synchronous motor also has many low-speed applications,such as low-speed electric vehicle drive system,industrial large-scale low-speed motor and so on.The performance of traditional control algorithm sometimes decreases at low speed.And the model-based sensorless algorithm used at medium and high speed can not be used at low speed.Based on the above problems,some research has been done in this paper about the sensorless model predictive control system of PMSM at low speed.Firstly,the principle of model predictive control is analyzed based on the mathematical model of PMSM in different coordinate systems.A continuous control set model predictive current control method is proposed to replace PI control.This method is equivalent to combining predictive control with vector control,and has the advantages of both methods.Compared with finite control set model predictive control,It increases the diversity of switching action selection of converter,and makes switching frequency constant.Compared with traditional vector control,it has high dynamic response ability.It can ensure that the output of the system follows the given rapidly,so it has good control performance at low speed.At the same time,according to the disadvantage that the method is greatly affected by parameter disturbance,the incremental state equation is proposed to design the controller,and the disturbance observer is designed to compensate the component of the mismatch model,so as to improve the robustness of the method to parameter disturbance.High performance PMSM control technology requires accurate rotor position information,which is generally measured by mechanical sensors.However,Because of mechanical sensors' characteristics of the high cost,large volume and poor anti-interference ability,it is necessary to use or reserve the position sensorless method to improve the reliability of the system.Compared with the model-based sensorless control method at medium and high speeds,the model-based method can not be used at low speeds because the back EMF value is very small.In this paper,an improved high-frequency square wave injection method is proposed,this method injected high-frequency square-wave voltage on the ?? axis respectively which decrease the use of filters or the calculation of DC bias in the s ignal processing,which greatly improves system bandwidth and improves estimation accuracy.The method is simulated by Matlab/Simulink software and verified by experiment on the test platform of permanent magnet synchronous motor with TMS320F2808 DSP as the control core.The simulation and experimental results show the effectiveness of the proposed method.