Research On Parameter Self-tuning Methods Of PI Controller For AC Servo System

Permanent magnet synchronous motor is the power component of the PCB processing CNC machine tools,its control performance directly affect the precision and efficiency of the processed components,the influence of the moment of inertia on the system performance is the most significant.To enhance the adaptability and robustness of the system and improve the performance of PCB processing CNC machine tools needs to online monitor the change of the moment of inertia of the motor,and dynamically adjust the PI parameters of the servo controller based on the moment of inertia,optimize the mechanical properties of the motor.The directional vector control scheme of rotor magnet field with zero direct axis current was determined for the strong coupling and nonlinear complex structural characteristics of permanent magnet synchronous motor.In this control scheme,to get high precision and fast convergence speed of identification value,on the basis of detailed analysis of basic principle of system identification,the motor parameters online identification strategy was studied.Firstly,the moment of inertia identification scheme based on least-squares was designed,by studying the influence of data saturation phenomenon of least square method,a novel control strategy combined forgetting factor weighted with median filter was put forward.Next the moment of inertia identification scheme based on extended Kalman filter was designed for real-time estimation of the moment of inertia when the rotational speed was in a rapid time-varying state disturbed by the outside.Finally,the convergence time and steady-state error of the output curve of two online identification algorithms were analyzed by simulation,the optimal value of identification results was selected as the basis for the servo controller parameter setting.Under the optimum system design method,it can be concluded that the linear proportional relationship between the moment of inertia and proportional gain of speed loop PI controller.For purpose of the system ability of fast response and strong robustness,the PI controller parameters of current loop and speed loop online tuning scheme was designed based on the moment of inertia identification value.In this paper,the small and medium power motor usually used in PCB processing was taken as an example to verify the feasibility of the overall scheme.The simulation results show that given speed was changed and external disturbance was added,the motor inertia can be identified online and the internal parameters of the controller can be adjusted adaptively,the AC permanent magnet synchronous servo system is always in good output characteristic.