Research On PMSM Control System Based On Low Resolution Position Sensor

Permanent Magnet Synchronous Motor(PMSM)is widely used in various industries due to its superior dynamic performance,light weight and high efficiency.With the development of power electronics technology and microelectronic control technology,it plays an important role in the modern AC motion control system in recent years.At present,vector control,as the most widely used PMSM control strategy,can achieve precise control of motor speed and current.Vector control needs to know the accurate rotor position signal,so PMSM rotor position detection technology is very important.The use of high-precision position sensors can ensure that high-precision rotor position signals are obtained,but it also increases the space and cost of the system,and is limited by the environment,reducing reliability;using sensorless algorithms can greatly reduce costs,But its algorithm is more complicated and sensitive to motor parameters.In order to reduce the cost of the PMSM control system and obtain high-resolution speed and rotor position information,this paper uses a low-cost,low-resolution Hall position sensor to achieve efficient and reliable rotor position estimation.This article first introduces the research status of PMSM low-resolution position sensor control technology.In addition,the mathematical models of PMSM in three coordinate systems are introduced,and the basic principles and specific implementation of PMSM vector control technology and space vector pulse width modulation(Space Vector Pulse Width Modulation,SVPWM)algorithm are explained.Secondly,the installation of the switch-type Hall position sensor and the position signal of the Hall switch are analyzed,and the method of estimating the motor speed and rotor position by detecting the time interval of the Hall interval is introduced.The estimation principles of the zero-order algorithm and the first-order algorithm are described in detail,and on this basis,a first-order algorithm for q-axis current compensation is designed to improve the performance of the motor at startup or at low speed.A surface-mount PMSM vector control simulation model was built in Matlab software to verify the zero-order algorithm,the first-order algorithm and the first-order algorithm of q-axis current compensation,and the effectiveness of the three estimation methods was verified.The speed and acceleration used by the three estimation methods are derived from the previous interval,and there is a certain degree of lag in the estimation results when the motor accelerates and decelerates.In order to further reduce the estimation error of the motor speed and rotor position without increasing the cost of the PMSM control system,and to obtain high-resolution speed and rotor position information,a new estimation method combined with the first-order current compensation algorithm is designed.When the PMSM is running,a set of speed and rotor position data is obtained through the first-order algorithm of current compensation;at the same time,the three-way Hall signal output by the Hall sensor is subjected to coordinate transformation and a designed adaptive filter to obtain the rotor position information.The fundamental component of the Hall vector is used to extract another set of speed and rotor position data using a quadrature phase locked loop(QPLL).Then,the weighted average processing method is used for the two sets of data to obtain the final estimated speed and rotor position.A simulation model was built in Matlab to verify the feasibility and effectiveness of the new estimation method.Finally,in order to verify the correctness of the above control methods,STM32F103ZET6 is used as the control chip to build a PMSM experimental platform,and the first-order algorithm and the designed new estimation method are experimentally verified and compared.The experimental results show that the new estimation method based on the Hall position sensor designed in this paper can obtain high-resolution speed and rotor position signals.When the motor is started,there is only a maximum angle error of about 0.35 rad.The maximum error of position estimation is about 0.1 rad,which meets the requirements of PMSM high-performance drive control.