Dynamic Modeling and Control of Atomic Force Microscope probe measurement system

• Main Authors: Shou-Feng Chiu, 邱碩峰
• Other Authors: Chung-Feng Jeffrey Kuo
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/xc355v

碩士 === 國立臺灣科技大學 === 高分子系 === 94 === The objective of thesis is to derive the mathematical model and design the controller of the micro-cantilever probe measurement system of atomic force microscopy. However, most of the distributed parameter control system for atomic force microscopy are based on reduced-order models, but the problems of result are often in “ computation errors” or “control and observation spillover”. Therefore, developing some realizable controllers which cannot only stabilize all the vibration modes but also make such system efficient for good tracking is crucial. First, the governing equations of motion and associated boundary conditions are derived from Hamilton’s principle. Using the separation of variables method, the dynamic characteristics such as natural frequencies and mode shapes function of the system can be analyzed. And then, the corresponding open loop transfer function will be obtained from Laplace transform. Finally, the actuator and sensor will be designed from collocated control. The results show that the poles and zeros patterns will interlace along the imaginary axis in the root locus and have characteristic of minimum phase property. It is the essential characteristics for designing the controller for such system. According to root locus method the tracking property in the closed loop system can be analyzed. Avoid all the poles and zeros cancellation with each other to influence the system to be steady, then the augmented root locus method is used to design the controllers such as Proportional-Derivative controller and Phase Lead compensator. It can be shown that system cannot only eliminate infinite dimensional vibration modes but also make system stable. Finally, the objective function of Genetic Algorithms can be designed for to find out the best parameter of controller in order to reduce computational process and improve accuracy. It can be seen from computer simulation, the designed control system for atomic force microscopy can accord with performance index to reach precision control.