Design Of Permanent Magnet Synchronous Motor Control System Based On Time Grating Position Feedback

Permanent magnet synchronous motor(Permanent Magnet Synchronous Motor:PMSM)has been widely used in many fields because of its simple structure and good output characteristics.When the Field-oriented Control(FOC)algorithm controls the PMSM,it has the advantages of low noise,stepless speed regulation,fast dynamic response and strong shock load resistance.In addition,FOC can also control the PMSM as a DC motor.,which simplifies the control process of PMSM,so the FOC algorithm is the most widely used among the various control algorithms of PMSM.However,it is difficult to adjust the parameters of the PID part of the FOC to the optimum,so this topic uses a neural network to improve the PID part.The improved algorithm is easy to obtain the local optimal parameters of the PID part,that is,the neural network PID-FOC control algorithm.The neural network PID-FOC control algorithm needs to measure the position information of the motor rotor in real time,and the photoelectric encoder has the disadvantages of high cost,difficult installation,poor anti-shock and interference ability,so the use of strong antiinterference ability,high measurement accuracy and resolution A time-gated displacement sensor with high cost,low manufacturing cost and strong environmental adaptability is used to perform position feedback on the motor rotor.In this paper,the time grid displacement sensor is used as the closed-loop position feedback of the control system,and the control system of PMSM motor is designed by using the neural network PID-FOC control algorithm.Mainly complete the following work:(1)Algorithm design of PMSM control system: The neural network PID-FOC control algorithm mainly includes three parts: coordinate transformation,space vector pulse width modulation(SVPWM)and neural network PID.First analyze and design each module separately,and then integrate the three modules into a neural network PID-FOC algorithm.The coordinate transformation module mainly decouples the strong coupling characteristics of the PMSM,which converts the three-phase alternating current that controls the PMSM into two mutually orthogonal and independent direct currents,which are the torque current(Iq)and the excitation current(Id);SVPWM mainly controls the output of the inverter.SVPWM controls the PMSM by controlling the switching state of the six MOSFET tubes in the inverter,so that the inverter outputs a six-beat staircase wave similar to the sinusoidal alternating current;the neural network PID is mainly through The neural network algorithm is used to optimize the PID part of the FOC algorithm,so that it can easily obtain the local optimal PID parameters,so as to optimize the FOC control algorithm,so that the motor has a faster response speed and the ability to resist shock loads.(2)PMSM control system algorithm simulation: model simulation of neural network PID-FOC algorithm.The model of PMSM control system is built under Simulink,and the simulation experiment is carried out.(3)PMSM rotor position detection: When using the neural network PID-FOC control algorithm to control the PMSM,in order to achieve high-performance control of the PMSM,the motor rotor position must be detected in real time.In this paper,the time grid displacement sensor will be used to detect the position information of the motor rotor in real time.The time grid sensor is isochronous sampling,while the photoelectric encoder is used to measure the spatial position of the code disc with equal spacing and engraving.The two have different working principles.,if the time grid sensor is directly matched with the PMSM control system,the sampling period and sampling space will be out of sync.Therefore,a prediction measurement method based on BP neural network is proposed to solve the mismatch between the time grid and the PMSM control system.question.(4)Software and hardware design and verification of PMSM control system: According to the model designed in Chapter 2,the control program is generated and modified,and the generated C program is modified,integrated and debugged in Keil software,so that the control of the motor can achieve the desired result.The core circuit of the controller includes:STM32 minimum system,power module,three-phase inverter drive module,current sampling,communication module,hardware overcurrent protection and other modules;controller design includes circuit diagram design and PCB electrical system design,and according to specific applications Object and control requirements for component selection.Experiments show that the neural network PID-FOC control system based on the time grid position sensor has good control performance for PMSM at medium and low speeds,so the control system can meet the control requirements of the experimental turntable for PMSM.