Research On Control Algorithm Of Permanent Magnet Synchronous Motor Based On Improved Sliding Mode Observer

With the continuous development of permanent magnet materials and the process of intelligent industrialization in China,permanent magnet synchronous motors（PMSM）have been widely used in various industries because of their high efficiency,high control precision,and significant power density.The permanent magnet synchronous motor control algorithm has also become a research hotspot.The traditional sensing control algorithm relies on the position sensor,which is easy to be affected by the accuracy of the sensor and the surrounding working environment,and additionally increases the use cost of a permanent magnet synchronous motor.Therefore,the research on the control algorithm without a position sensor can replace the sensor,which has rich application value and research significance.This thesis takes the surface-mounted permanent magnet synchronous motor as the research object,analyzes the motor vector control method,and based on the traditional sliding mode observer,for chattering,low control accuracy,rotor phase delay,and other problems,proposed an improved sliding mode observer,optimize the performance of the traditional sliding mode observer,to achieve high-performance sensorless control of permanent magnet synchronous motor.Firstly,this thesis introduces the basic structure and operation principle of PMSM,deduces the mathematical formula model of PMSM under different reference coordinate systems using the coordinate transformation method,and then analyzes and deduces the basic idea and specific principle of vector control strategy,and studies the space vector pulse width modulation technology and sliding mode variable structure control theory.And a sensorless vector control simulation model is established based on the traditional sliding mode observer.Secondly,the chattering,low observation accuracy of the traditional sliding mode observer and rotor phase delay are optimized.In order to improve the traditional sliding mode observer,the switching function is replaced by a new quasi-sliding mode function to ensure the robustness and suppress the chattering of the motor.A new convergence law is designed by introducing the exponential term of the system state change,and the fixed sliding mode gain is improved to the speed adaptive gain which improves the dynamic performance of the motor.In the rotor position and speed estimation module,the phase-locked loop is used to replace the traditional arctangent function,which improves the observation accuracy of the motor.Eventually,an improved sliding mode observer is designed,the speed error is reduced by 75%,the delay time of the rotor position is reduced by 0.15 ms and effectively reduces the chattering situation.Finally,to verify the effectiveness of the improved sliding mode observer in this thesis,a hardware module based on FPGA is designed using the System Generator.The control system’s PI controller module,coordinate transformation module,SVPWM module,improved sliding mode observer module,and rotor position and speed estimate module are designed.The CORDIC algorithm is used to calculate the nonlinear function,and the corresponding IP core is generated and downloaded to the FPGA experimental platform,Finally,the hardware-in-the-loop simulation experiment was carried out jointly with MATLAB/Simulink platform to verify the effectiveness of the improved sliding mode observer.The experimental results show that the improved sliding mode observer reduces the speed error range from±20r·min^-1 to±5r·min^-1,the speed estimation accuracy is increased by 75%,the rotor phase observation delay is reduced from 0.2 ms to 0.05 ms,and the delay time is reduced by 75%.The effectiveness of the improved sliding mode observer is verified.