Based On Adaptive Iterative Feedforward And Interference Suppression Robust Tracking Control Of Wafer Stage

Lithography machine is an important manufacturing equipment in the process of semiconductor production.As an important part of the lithography machine,the motion control precision of the workpiece table directly affects the efficiency and quality of semiconductor production.In the process of lithography production,the workpiece table needs to realize fast tracking in the step movement stage and precise positioning in the exposure stage.However,it is difficult to give consideration to the tracking speed and positioning accuracy due to the constraints of actuators,the large inertia of workpiece platform and the influence of disturbance of workpiece platform.To solve the above problems,this paper conducts in-depth research based on the control methods of iterative feedforward controller,fuzzy parameter adaptive adjustment,iterative learning and disturbance observer.Specific research contents are as follows:(1)The overall structure,components and working principle of the motion control system of the workpiece table are studied,and the mechanism model of the workpiece table is studied and the sweep frequency experiment is completed.The transfer function of the workpiece table motion control system is obtained by the method of function fitting,which provides a basis for the design of the subsequent controller.The motion trajectories of workpiece table at different transition times and the control problems under the motion trajectories are discussed.The disturbance of workpiece table is analyzed and tested.(2)The principle and characteristics of the iterative feedforward controller are analyzed,and the high order variable gain iterative factor is adopted to improve the robustness of the system and the speed of iterative convergence.In order to solve the problem of system parameter perturbation and control in different moving stages,fuzzy control strategy is introduced to adjust the parameters of feedforward controller adaptively based on iterative feedforward control,and fuzzy neural network is used to adjust the PID parameters of feedback controller.This method can effectively improve the tracking accuracy of transition stage,but the introduction of feedforward control quantity leads to the decrease of steady-state accuracy.To solve this problem,an optimal switching control strategy is introduced to realize the switching of feedforward control quantity,so as to avoid the deterioration of steady-state accuracy and system oscillation during the switching.(3)Aiming at the problem of disturbance affecting tracking control accuracy in the system,iterative learning was combined with the control strategy of disturbance observer,which was introduced into the control structure.Iterative learning was used to suppress repetitive disturbance,disturbance observer was used to estimate the non-repetitive disturbance of the system and compensate by feedback loop.The disturbance observer used sliding mode disturbance observer and extended state observer respectively.(4)In order to solve the problem of slow convergence of the sliding mode perturbation observer and chattering of the system,a new approach law containing system state variables is designed.Compared with the traditional approach law,it has a faster convergence rate and weakens chattering phenomenon.A cascade of extended state observer is used to design a nonlinear function and suppress the peak phenomenon.The experimental results of the workpiece table motion control system show that the proposed method can effectively suppress the repetitive disturbance and non-repetitive disturbance in the system and improve the robustness of the system.