Research On Learning Control Method Of Micro Nano Structure Imaging

Micro-nano structure imaging technology is a bridge between the study of the micro world and the macro world,and it is different from traditional nano imaging technology.The micro-nano structure imaging technology uses dual scanners of a micro-structure imaging platform and a nano-structure imaging platform for scanning and imaging.The structure is complex and the operation is cumbersome.It is necessary to select a suitable control method to obtain a clearer image.In order to improve the imaging quality of the micro-nano structure imaging system,this paper firstly analyzes the various components of the micro-nano structure imaging system,and then establishes the mathematical model of the micro-nano structure imaging system by using two methods:mechanism structure and experimental data.The learning control strategy is adopted to design open-loop and closed-loop PD-type iterative learning controllers to improve the control tracking accuracy of the micro-nano structure imaging system.Then,in view of the shortcomings of the open-loop learning controller only using the error information of the previous iteration and poor anti-disturbance,and the closed-loop learning controller only using the current iteration error information,an iterative feedforward feedback learning controller based on the open-closed-loop learning law is designed.;Aiming at the slower convergence speed in the learning process of open and closed loop iterative learning controllers,the traditional fixed learning gain is changed to exponentially variable gain to improve the convergence speed of the controller.Through the MATLAB/Simulink simulation experiment,the feasibility of the learning control method in the micro-nano structure imaging system is proved.Finally,through the analysis and comparison of the average gradient algorithm and the improved point sharpness definition algorithm,the effectiveness of the learning control method is verified.Imaging experiment results:in the scanning range of100×100?m and 40×40?m,the imaging clarity of the micro-nano structure imaging system under the control of feedforward feedback learning is 4.2633,1.4652×10~6 and10.1459,1.9534×10~7,respectively,indicating that iterative learning feedforward feedback control is compared with PID control and closed-loop PD type iterative learning control,the system imaging is clear and high,and the edge details are rich in information.The micro-nano structure imaging system learning control method studied in this paper can effectively improve the imaging quality of the system,greatly broaden the application range of the micro-nano structure imaging system,and open a new door for the learning control method in the field of micro-nano imaging technology.