Research On Parameters Identification Of PMSM Based On Sensorless Control

Permanent Magnet Synchronous Motor(PMSM)has merits of high efficiency,high power density,wide range speed adjustment and fast torque response,which has been widely used in industrial manufacturing,new energy vehicles,aerospace and other fields.Vector control is a common control method to achieve high performance operation of PMSM,but it needs to obtain the rotor position information in real time.The traditional way of installing mechanical position sensors increases the cost of the control system,takes up the system’s space,and reduces the system’s reliability.In order to solve these problems,sensorless control of PMSM has become one of the research hotspots of electronic control systems.During the operation of PMSM,its parameters such as the stator resistance and inductance are easily influenced by temperature rising of winding,magnetic saturation and other factors,which leads to the decline of system’s reliability and control performance.In order to realize the fault diagnosis of the PMSM sensorless control system and ensure the stability of the system,it is very necessary to identify the parameters of PMSM.This paper takes the surface-mounted PMSM as the research object to study sensorless vector control and parameter identification of PMSM.The main work of this paper is as follows:1.This paper adopts twice current injection to complete pre-orientation of motor’s rotor,study the motor low-speed starting strategy with I/F control,and use the given current regulator to make the motor from low-speed with I/F control switch to the medium and high-speed area smoothly with the speed-current double closed-loop control.When the motor is running at medium-high speed,the improved SMO is designed to estimate the rotor position of PMSM.Aiming at the chattering phenomenon caused by traditional SMO,the improved observer adopts saturation function instead of switching function to weaken the system chattering,and uses back electromotive force observer to calculate the back electromotive force of PMSM,which can avoid the phase shift caused by using low pass filter,then the phase-locked loop is used to improve the estimate accuracy of rotor position.Finally,the effectiveness of the improved sliding mode observer is verified by a simulation comparison with the traditional SMO.2.The offline and online identification of PMSM parameters is studied.The recursive equation of motor parameters identification based on RLS and the adaptive law of motor parameters identification base on MRAS are deduced.According to the advantages of RLS and MRAS,this paper adopts a parameter identification strategy with RLS-MRAS based on the sensorless control of PMSM.Firstly,the inductance of PMSM is identified by RLS,then takes the inductance and the rotor flux that estimated by SMO as the input of MRAS to identify the stator resistance,which achieve the multi-parameters identification of PMSM.Finally,the effectiveness of this algorithm is verified by simulating at motor load mutation,speed mutation,parameters mutation and other working conditions in the MATLAB/Simulink environment.3.This paper takes a blower as controlled object,and the experimental platform including hardware and software are designed based on STM32F103C8T6 main control chip.Stator resistance and inductance are identified in offline firstly,then the operation performance of the blower with sensorless vector control is tested,including low-speed performance with I/F control and medium-high speed performance with sliding mode observer to estimate the rotor position,and the online identification of blower’s parameters is researched under vector control.The experimental result shows that the parameter identification strategy can realize the online identification of the stator resistance,stator inductance and rotor flux of the blower under certain working condition.And the blower runs stable under sensorless control and has certain practical value.