Research And Implementation Of Position Sensorless Control Technology For Permanent Magnet Synchronous Motor In Full Speed Range

With breakthroughs in the research of permanent magnet materials and high withstand voltage power electronic devices,permanent magnet synchronous motors are increasingly used in high-precision control applications,such as aerospace,automation control,medical equipment,electric vehicles and other fields.In high-precision control systems,closed-loop feedback of speed and rotor position signals is essential.However,the traditional way of using mechanical sensors to obtain rotor position information has many disadvantages.How to extract the rotor position information from the feedback measurement signal of the motor without adding hardware equipment has become a current research hotspot.There is no single control strategy that can realize the position sensorless control of the permanent magnet synchronous motor in the full speed range.How to realize the position sensorless control in the full speed range has become a current research hotspot.This article focuses on the in-depth study of this issue,and the research content is as follows:1.Use the pulse vibration high frequency voltage injection method to realize the rotor position identification in the zero/low speed section of the motor.On the basis of completing its theoretical derivation,analyze the existing shortcomings.In order to increase the frequency of the injected voltage and reduce the number of filters used in the speed identification system,the injected high-frequency voltage is replaced from a sine wave to a square wave,and based on the current response characteristics of the square wave voltage excitation,a method suitable for motor stability and In the transient process,the filter-free high-frequency response current separation and filter-free position error signal extraction strategy are derived,and its applicable scope is deduced.On this basis,the magnetic circuit saturation effect is used to complete the magnetic identification of the rotor permanent magnet poles.This extraction method is combined with the PID-type Luenberger position observer to complete the rotor position identification.2.The model reference adaptive method(MRAS)is used to realize the rotor position identification in the middle/high speed section.On the basis of completing the theoretical derivation of MRAS,the existing deficiencies are analyzed and corresponding improvement strategies are proposed.Aiming at the problem that the MRAS speed identification system is sensitive to motor parameters,this paper proposes the least square method of motor parameters with forgetting factor to identify the motor inductance parameters online,and apply the identification value to the MRAS system to improve its speed identification accuracy.In view of the many problems caused by the PI controller as the adaptive rate of the MRAS speed identification system,the BP neural network is used as the adaptive rate of the MRAS system to construct the rotor position identification system,and the corresponding neural network learning rules are derived to realize the motor speed Identify quickly.On the basis of this improvement,the SVPWM output duty cycle is multiplied by the bus voltage to reconstruct the three opposite electromotive force at low speed of the motor,improve the low speed identification accuracy of the BP-MRAS speed identification system,and expand the rotor speed identification range of the MRAS system.3.Select the weighting method to complete the transition of the two identification algorithms,and realize the permanent magnet synchronous motor’s full speed range without position sensing control operation.4.According to the requirements of high-frequency voltage injection method for current sampling accuracy,improve the current sampling hardware part of the original d SPACE motor test bench,reselect the current sensor and complete the design of its corresponding signal processing circuit.Relevant implementation verification is completed on this experimental platform.The experimental results verify that the proposed IPMSM speed identification algorithm has better identification accuracy and anti-jamming performance.