基於響應預測之多重控制器切換及其於精密定位平台之應用

Multiple Switching Control based on Response Prediction：with Application to a Long-Stroke Precision Stage

陸俊甫(Jun-Fu Lu)

指導教授 ： 王富正

國立臺灣大學/工學院/機械工程學研究所/碩士(2020年)

https://doi.org/10.6342/NTU202002170

摘要

本論文發展一套多重控制器切換機制，藉由預測系統未來響應決定控制器最佳切換時機，並將其應用於大行程精密定位平台，結合雙光子聚合技術製作微結構物，以光學性質顯示所提出的控制器切換機制確實有助於提升微結構物製作精度。
隨著科技快速發展，許多高科技產業與產品走向精密化、微小化的趨勢，例如：半導體產業製程、微機電系統製程、雙光子製程等等。其中壓電材料由於具有高精密度與響應快等優點已被廣泛的應用在精密定位系統，然而受限於材料的行程，使其無法達到大尺度的製造，因此本論文整合精密壓電平台及大行程步進馬達平台，其中壓電平台採用壓電材料進行精密定位，步進馬達可以增加平台整體行程，將其整合為大行程且高精密度之整合平台。
首先，我們針對壓電平台設計多個不同特性的強韌控制器，並藉由粒子群演算法將控制器降階，而為了結合多個控制器的優點，本論文提出多重控制器切換架構，預測壓電平台未來響應來決定控制器使用的優先順序，來達到最佳的追跡響應。其次，我們針對步進馬達平台設計前饋控制器並結合增益調變比例控制器，前饋控制器可以減少追跡誤差與相位落後，而增益調變控制器可以調配馬達的速度，來增進馬達平台的追跡能力。最後我們整合兩種平台並提出雙迴圈控制架構，透過預測整合平台的輸出響應來決定壓電平台控制器的切換順序，修正整合平台的誤差，最終以模擬與實驗展現大行程且精密的定位能力。
我們進一步將多重控制器切換機制應用於雙光子聚合製程，分別製作直徑130μm的微透鏡結構物與刻畫長度205μm且寬度30μm的文字結構物，以SEM拍攝微結構物成品與測試微透鏡的光學性質，顯示所提出的多重控制器切換機制有助於提升平台精密定位性能。

關鍵字

壓電材料 ； 步進馬達 ； 精密定位平台 ； 強韌控制 ； 粒子群演算法 ； 控制器切換 ； 雙光子聚合

並列摘要

This thesis develops a multiple controller switching mechanism for a long-stroke precision positioning stage. This mechanism determines the optimal control switching sequences by predicting the future response. We further integrate the precision positioning stage with a two-photon polymerization (TPP) system to fabricate micro-structures. The optical properties of microstructures show that the proposed controller switching mechanism is effective in improving of microstructure fabrication.
With the advance of technology, precision positioning techniques are becoming more amd more important for high-tech industries, such as semiconductor manufacturing process, microelectromechanical systems and two-photon manufacturing process. Piezoelectric transducer (PZT) is usually applied for precision positioning because of its fast response and high resolution. However, travel distance of PZT is limited because of the material properties. Therefore, we integrate the the PZT stage with a stepper motor stage to achieve high precision with long stroke.
First, we design several controllers with different advantages for the PZT stage and propose the multiple switching control architecture to predict the future response and to switch the controllers for improving tracking response. Second, we design a feedforward controller and a gain scheduling controller for the stepper motor stage. The feedforward controller can reduce tracking errors, while the gain scheduling controller can adjust the speed of the motor to improve the tracking ability. Third, we integrate the two stages and propose a double-loop control architecture, so that the PZT stage can compensate the errors caused by the stepper motor stage. Finally, we demonstrate the long-stroke precise positioning capability through simulation and experiment.
We further integrat the combined stage with a TPP system to fabricate microlens with a diameter of 130 μm by Fresnel zone plate (FZP) and a text structures with a length of 205 μm and a width of 30 μm. We test the optical properties of the microlens and observe the structure by SEM. The results confirm the precise positioning performance of the combined stage employing the proposed switching control.

並列關鍵字

Piezoelectric material ； Stepper motor ； Precision positioning stage ； Robust control ； Switching control ； Two photon polymerization

參考文獻

[1] S. Mishra, L. Unnikrishnan, S. K. Nayak, S. Mohanty. Advances in Piezoelectric Polymer Composites for Energy Harvesting Applications: A Systematic Review. Macromolecular Materials and Engineering 2019；Vol. 304, Iss. 1.

[2] J. Gan, X. Zhang. Nonlinear Hysteresis Modeling of Piezoelectric Actuators Using a Generalized Bouc–Wen Model. Micromachines 2019；Vol. 10, Iss. 3, Pgs. 183.

[3] G. Chen, X. Yan, J. Cai, H. Guo. Hysteresis nonlinear modeling andcompensation of piezoelectric ceramic sensors in micro measurement systems. Measurement Science and Technology 2018；Vol. 29, Num. 9.

[4] J. Fang, J. Wang, C. Li, W. Zhong, Z. Long. A Compound Control Based on the Piezo-Actuated Stage with Bouc–Wen Model. Micromachines 2019; Vol. 10, Iss. 12, Pgs. 861.

[5] Z. Wei, B. L. Xiang, R. X. Ting. Online parameter identification of the asymmetrical Bouc–Wenmodel for piezoelectric actuators. Precision Engineering 2014；Vol. 38, Iss. 4, Pgs. 921-927.