基於交叉耦合分數階自抗擾控制之X-Y-Y棒狀線性馬達定位平台

Abstract

為了能夠使設備的追蹤效果以及動子間的同動性能提升，本論文設計出交叉耦合分數階自抗擾控制(CCFOADRC)策略，用於控制X-Y-Y棒狀線性馬達定位平台。首先介紹棒狀線性馬達平台之系統架構和運作原理，通過時域的系統鑑別推導出馬達數學模型中的系統參數。接著，設計出第一個控制器為自抗擾控制器(ADRC)，在模擬確認能良好做出控制之後，為了更進一步改善棒狀線性馬達的定位誤差及同為Y軸的定位誤差相減產生的同動誤差，針對定位誤差的改善加入了分數階微積分做改善，設計出了分數階自抗擾控制器(FADRC)，通過了分數階提供的額外自由度，成功的改善其控制響應，接著為了改善同動誤差，加入了交叉耦合控制，進一步提出了交叉耦合分數階自抗擾控制器(CFADRC)。交叉耦合分數階自抗擾控制器裡包含了許多控制項，複雜度也隨之升高，因此本論文提出了智慧型交叉耦合分數階自抗擾控制器(ICFADRC)，藉由教與學最佳化方法(TLBO)針對重要參數做動態優化。在教與學最佳化方法的過程中，進一步引進灰狼演算法的概念設計出助教型教與學演算法(TA-TLBO)。最後，由實作結果可以得知本論文提出的控制策略能有效地控制X-Y-Y棒狀線性馬達定位平台。In order to improve performance of tracking and co-movement between movers, this thesis designed a cross-coupled fractional-order active disturbance rejection control (CCFOADRC) strategy to control the X-Y-Y tubular linear motor positioning positioning stage. First, the structures and dynamics of the X-Y-Y tubular linear motor positioning stage are described. Then, the active disturbance rejection control (ADRC) controller will be introduced. After the simulation using MATLAB confirms that it has well performance control, this study is willing to further reduce the positioning error of the tubular linear motor and the synchronized error caused by the subtraction of the positioning error of the other Y axis. To improve the positioning performance, fractional calculus was added to the controller, and the fractional active disturbance rejection controller (FADRC) was designed. Through the additional degrees of freedom provided by the fractional order, it successfully reduced its control response. Then, in order to reduce the synchronized error, cross-coupling control was added, and a cross-coupling fractional-order active disturbance rejection controller (CFADRC) was designed. Due to the cross-coupled fractional-order ADRC contains many control items, the complexity also increases. Therefore, this thesis proposes an intelligent cross-coupled fractional ADRC (ICFADRC), which is optimized by teaching and learning optimization (TLBO) method for dynamic optimization of important parameters. In the process of teaching and learning optimization method, the concept of gray wolf algorithm was further introduced to design the teaching assistant-type teaching and learning algorithm (TA-TLBO). Finally, the results show that the proposed control strategy can control the X-Y-Y tubular linear motor positioning stage effectively.