MFAILC-Based Precision Motion Control Algorithm For Wafer Stage

Lithography has become a decisive semiconductor manufacturing equipment.As one of the most important components of lithography,the motion accuracy of wafer stage determines the manufacturing accuracy of semiconductor chip.The wafer stage of the stepper lithography needs to maintain a rapid transition in the stepping process and accurate positioning in the exposure process.However,due to the constraints of the actuator,the large inertia of the wafer stage,the disturbance and internal noise of the wafer stage,it is difficult to achieve rapid transition and high-precision positioning at the same time.Aiming at the problems of rapid and precise positioning of wafer stage under the above disturbances and constraints,thesis studies the trajectory tracking control method of wafer stage based on nonlinear auto disturbance rejection control and model free adaptive iterative learning control.The specific research contents are as follows:1)The composition structure and working principle of the wafer stage of a step projection lithography motion control system are studied,and the disturbance of the wafer stage is analyzed and measured.Then,the trajectory of the wafer stage is planned,the trajectory tracking control problems of wafer stage system under the influence of disturbance is summarized under the conditions of trajectory invariance and trajectory change,and the model of the wafer stage system is identified through the frequency experiment as reference for controller design.2)Aiming at the control problem that the wafer stage system is affected by internal and external disturbances,a feedforward feedback switching control method based on model-free adaptive iterative learning is proposed.Firstly,the nonlinear ADRC is designed as the position loop controller to compensate the nonrepetitive disturbance of the wafer stage.Then,a model free adaptive iterative learning feedforward controller is designed to compensate the repetitive disturbance of the system,and considering the motion characteristics of the wafer stage,the model free adaptive iterative learning controller is improved.An optimized switching control method is used to switch the two iterative learning controllers.It is verified on the wafer stage experimental system of lithography machine,and shows that compared with the traditional control method,this control method can greatly improve the dynamic performance of the system,and has better steady-state positioning accuracy,but the convergence speed of iterative learning is slow.3)In order to improve the disturbance suppression ability and convergence speed of the algorithm,a high-order model free adaptive iterative learning algorithm with disturbance compensation is proposed.Firstly,a model free adaptive iterative learning control method with disturbance compensation is designed,and then a high-order model free adaptive iterative learning control method with disturbance compensation is designed.The verification on the wafer stage experimental system shows that this method has faster convergence speed and significantly improved dynamic performance.4)Aiming at the problem of poor convergence of the control method based on model free adaptive iterative learning control under the change of reference trajectory of wafer stage,a model free adaptive iterative learning algorithm with a variable forgetting factor is proposed.Thesis designs a model free adaptive iterative learning control method with variable forgetting factor,and then a feedforward switching method is designed to switch between algorithms with and without forgetting factor to improve the servo ability of the algorithm.Finally,it is verified on the wafer stage experimental system of lithography machine,shows that this method balances the extrapolation property and servo ability,and can realize faster tracking of the new reference trajectory when the reference trajectory of wafer stage changes.