Research On The Motion Control Technology Of The Short-stroke Stage Of The Positioning Stage System In Lithography Machine

The short-stroke stage is one of the key components of the lithography machine's positioning stage system.Its internal mechanism and structure are very complex.At the same time,the short-stroke stage has a small stroke and high precision requirements.Its positioning accuracy and movement accuracy will directly affect the wafer in the light The quality of the engraving process.In this paper,the research on the motion control of the short-stroke stage of the lithography machine's positioning stage system is carried out,and the control parameter tuning method of the short-stroke stage is proposed to enable it to achieve higher motion and positioning accuracy.Detailed analysis of the performance and own characteristics of the short-stroke stage,and a controller design,to achieve the motion control index of the short-stroke stage.Through the model identification method,the 6-DOF model of the shortstroke stage was identified in the frequency domain.The model was identified for each degree of freedom of the six-degree-of-freedom short-stroke platform through experiments,and the frequency domain characteristic curve of60?1500Hz was obtained.The transfer function of each degree of freedom is obtained by fitting.The identification experiment adopts the model decomposition method under closed-loop conditions,uses multi-sine signals as excitation signals,analyzes the identification results,and combines the characteristics of the model to decompose the respective degrees of the micromotion stage into double integration links,delay links and high-order structural flexibility links.The calculation and analysis of these links have resulted in a more accurate model of their respective degrees,which provides a basis for the design of the motion controller of the short-stroke stage.Using the precise model of each degree obtained by the identification test,the sisotool toolbox of Matlab is used to design the motion controller of each degree.Two controller structures are designed,including: PID and lead-lag controller.The PID controller can achieve control for all 6 degrees of freedom of the short-stroke stage,but it is difficult for the control of a few degrees of freedom to meet the requirements of the control index.Therefore,this paper also designs a lead-lag controller to comprehensively improve the control performance of the system.In order to further improve the control performance,research has been carried out on the parameter tuning of the controller.As the neural network has super non-linear fitting and fast optimization capabilities in high-precision nonlinear control systems,the paper builds a neural network for controller parameter optimization,based on the BP neural network's gradient estimation of the objective function,The tuning of the controller parameters no longer depends on experience,and the convergence speed of the control error and the accuracy of the parameters are also improved,and the control effect of the controller is optimized accordingly.On the positioning stage system of the lithography machine,the positioning experiment and the trajectory tracking experiment were verified for the shortstroke stage.By comparing and analyzing the results of the two experiments,and then comparing with the previous experimental results,the feasibility and effectiveness of the research in this paper are verified.