摘要
伺服系統中，摩擦力常具有高度的非線性，其特性隨著時間、溫度與環境的不同而改變，故鑑定出精確模型相當困難，也因此摩擦力造成控制精度下降並產生許多問題。由於摩擦力可能隨時間變化且未知其邊界，故一般的強健控制器與適應控制器都無法設計。本文提出一個以函數近似法(Function Approximation Technique, FAT)為基礎的適應性補償器，期望將摩擦力消除。並透過Lyapunov分析法證明閉迴路系統穩定度，再以實驗驗證其可行性與性能。
關鍵字
無資料並列摘要
For servo positioning, consideration of the friction effect is very important when high precision is required. Friction not only reduces positioning accuracy but also induces lots of problems. Static lubrication is able to partially solve some of the problems, while dynamic compensation can give better solutions in general. Since, in most of the manufacturing applications, the friction effect cannot be precisely modeled and its actual value might be time dependent with unknown variation bounds, traditional robust designs or adaptive controls do not properly apply. In this paper, a function approximation based adaptive compensator is found to cover the friction effect in a servo system to improve positioning accuracy. Rigorous mathematical proof based on the Lyapunov-like technique is employed to justify closed loop stability and boundedness of internal variables. An experimental setup is constructed to evaluate the performance of the proposed strategy.
延伸閱讀
- Yang, J. H., & Huang, J. T. (2010). 適應性傅立葉類神經網路之摩擦補償器設計. 華岡工程學報, (25), 31-36. https://doi.org/10.29987/CCUHKJE.201001.0005
- 鍾孟良（2011）。摩擦力鑑別與前饋型補償方法研究〔碩士論文，國立臺北科技大學〕。華藝線上圖書館。https://www.airitilibrary.com/Article/Detail?DocID=U0006-1708201113494400
- 呂純怡（2004）。動態摩擦帶電機制應用於流量感測之研究〔碩士論文，崑山科技大學〕。華藝線上圖書館。https://doi.org/10.6828/KSU.2004.00023