Precision Servo Control System Based On An Extended State Observer

The AC servo drive technology has been widely used in daily life and industrial production areas,such as CNC machine tools,industrial robots,etc.In an actual control system,there exist various disturbance factors,such as the friction torque,the perturbations of parameter,the change of load torque,the modeling errors and the measurement noises.These disturbance factors will inevitably affect the performance of servo systems.So it is of great significance to explore for a composite control scheme with disturbance compensation mechanism.For a second-order servo system with integral and inertia parts,the thesis proposes a discrete-time robust composite nonlinear control(RCNC)approach with disturbance compensation,which can achieve accurate position control.The approach lumps the load disturbance,friction torque and the parameter perturbation,etc into an additional state variable,to be included in the system model,and extended state observer is then designed based on the augmented model to estimate and compensate the disturbance.By using the observer,the accuracy of servo system is improved.A linear control law based on disturbance feedforward compensation is adopted to achieve a fast and accurate tracking control performance,and a nonlinear feedback action is added to modulate the damping ratio of the closed-loop system and improve the transient performance.Disturbance in an actual system is usually time-varying.To estimate this kind of disturbance,the thesis designs a third-order reduced-order state observer,which can estimate and compensate the disturbance more accurately.In the design,the change of the disturbance within a short period is no longer restricted to be zero.A more accurate function for the disturbance is used instead,and a more accurate control effect is achieved.Secondly,by utilizing the mathematical model of the surface-type permanent magnet synchronous motor under dq coordinate system,the space state expression of position servo system is constructed,and a discrete-time robust composite nonlinear controller with more accurate disturbance compensation is designed.Specifically,the position loop and speed loop are treated as a mechanical subsystem with one loop.Vector control strategy of permanent magnet synchronous motor(PMSM)with id=0,and the cascaded control structure of the position loop and current loop for position servo system are adopted.A simulation model is built with MATLAB/Simulink software.Based on the identified parameters,a RCNC controller is designed for position control and simulation is conducted.The control performance of discrete-time RCNC control scheme under different target positions,load torques and parameter perturbation are analyzed.The simulation results show that with the proposed scheme,the transient performance and steady state accuracy are better,and the ability of disturbance rejection is improved,meanwhile the robustness against parameter changes is stronger.An identification module is built on the model,and the least square method is adopted to identify the system parameters,the result is then verified by simulation.Relative identification experiment is conducted,system parameters are identified in the experiment.Finally,the corresponding control algorithm is implemented with the CCS3.3 software,and experiments are carried out on a PMSM experimental platform with a TMS320F28335 DSC board.By comparing RCNC controller with other controllers,the thesis analyzes the advantages of the RCNC scheme.The performance of RCNC controller based on identified parameter is verified under different conditions of load torque and position target.The results show that discrete-time RCNC scheme can track a target position accurately and quickly,and a good control performance is achieved.