Sensorless Control Of Permanent Magnet Synchronous Motor Based On Model Predictive Control

Permanent Magnet Synchronous Motor（PMSM）has attracted extensive attention in industry due to its advantages of high efficiency,high power density and wide speed range.Model predictive control（MPC）,as a new control strategy,has the advantages of simple principle and easy handling of nonlinear constraints,attracting the attention of many researchers.At the same time,the sensorless PMSM drive control system requires no position sensor,which has the advantages of low cost,high reliability and no installation space,and further improves the compactness and power density of the motor drive system.However,sensorless drive control system can not meet the requirements of high precision and high dynamic response speed applications,and its comprehensive performance needs to be further improved.Therefore,based on the finite set model of permanent magnet synchronous motor to predict current control,this paper studies the key problems that restrict the performance of sensorless control system,in order to improve the control performance of PMSM sensorless drive system in the full speed domain.This paper first introduces the basic structure and mathematical model of permanent magnet synchronous motor,and analyzes the basic principle of predictive control of permanent magnet synchronous motor model based on this.Compared with traditional PI control,this method has the advantages of fast dynamic response and avoids the parameter adjustment brought by PI controller.At the same time,the working principle of SVPWM modulation is briefly described.Followed by detailed analysis of basic principle of sliding mode variable structure is designed for high speed rotor speed and position detection of the sliding mode observer,set out from two aspects,one is to eliminate and weaken the chattering: one is the extension of the phase-locked loop and the sliding mode observer is combined,to extract the rotor position signal,weaken the high frequency jitter estimated counter electromotive force component of the impact on the system is measured;Second,a high order superhelical sliding mode observer was designed.The switching function of sliding mode control was combined with the high order derivatives of sliding mode variables,which not only kept the advantages of invariance of traditional sliding mode interference,but also suppressed chattering and improved the control accuracy of the system.Low-speed conditions due to the fundamental wave mathematical model of the counter electromotive force value is very small,so often use high frequency injection method to extract the rotor position,this paper proposes a modified high frequency square wave injection method is adopted,this method in alpha beta shaft into different amplitude and phase difference of the square wave signals,using adjacent interval four kinds of high-frequency response current difference,The dc component of high frequency sines and cosines signal is eliminated by mathematical operation.Compared with the traditional high frequency injection method,the influence of filter on system bandwidth is reduced,and the error caused by DC bias on signal extraction is reduced,so as to improve the accuracy of position estimation.Finally,based on TMS320F28379 D three-phase PMSM drive control platform as the control core,experimental verification of the proposed method is carried out.Experimental results show that the proposed method can effectively improve the control performance of the motor drive system in the full speed domain,and have good anti-interference and dynamic response ability in the high-speed domain,and the chattering of the system is suppressed.At the same time,rotor position information estimation is accurate and speed tracking is stable in low speed domain.