發(fā)布時間：2018-06-18 23:25

  本文選題：切換系統(tǒng) + 線性變參數(shù)��； 參考：《哈爾濱理工大學(xué)》2017年碩士論文

【摘要】：在控制領(lǐng)域中,隨著技術(shù)的飛速發(fā)展和控制質(zhì)量要求的不斷提高,單純地依賴某一個系統(tǒng)或某一種系統(tǒng)的控制方式已經(jīng)無法滿足控制需要,混雜系統(tǒng)的提出使得這一問題得以解決。而切換系統(tǒng)就是一種典型的混雜系統(tǒng),簡單而言,切換系統(tǒng)就是由一系列的子系統(tǒng)和協(xié)調(diào)這些子系統(tǒng)的切換邏輯組成的。在具有不同需求的區(qū)域采用不同的控制方式以求達(dá)到最優(yōu)的控制效果,為了使系統(tǒng)運行能夠穩(wěn)定,控制系統(tǒng)的切換控制邏輯也成為了重中之重。近些年來隨著線性變參數(shù)(Linear Parameter-Varying,簡稱LPV)系統(tǒng)這一非線性系統(tǒng)在實際生產(chǎn)中的廣泛應(yīng)用,其存在的問題也慢慢地顯現(xiàn)出來。在LPV系統(tǒng)的設(shè)計中,往往由于參數(shù)范圍過大使得所設(shè)計的LPV系統(tǒng)在參數(shù)空間內(nèi)不能處處保證良好的控制效果。本文提出了一種關(guān)于線性變參數(shù)系統(tǒng)的切換控制方法,將調(diào)度參數(shù)的變化區(qū)間劃分成多個具有重疊區(qū)域的子區(qū)間,在這些子空間內(nèi)分別進(jìn)行控制器的設(shè)計,而重疊部分的控制輸出則由相鄰的兩個控制器共同決定。兩個相鄰的子系統(tǒng)在重疊區(qū)域同時運行,其運行結(jié)果經(jīng)過加權(quán)函數(shù)的運算后轉(zhuǎn)化為系統(tǒng)輸出。當(dāng)調(diào)度參數(shù)隨著時間的變化而進(jìn)入相應(yīng)的子區(qū)間時,調(diào)用相應(yīng)的子系統(tǒng)。同時為了降低所設(shè)計系統(tǒng)的保守性,在平均駐留時間內(nèi)的切換次數(shù)也將考慮在內(nèi)。本文所做的工作主要如下:首先,對于給定的參數(shù)變化區(qū)間,設(shè)計一簇子LPV系統(tǒng)狀態(tài)反饋控制器,保證其閉環(huán)穩(wěn)定且滿足給定的魯棒H∞性能指標(biāo)。參數(shù)依賴的Lyapunov函數(shù)矩陣與系統(tǒng)的參數(shù)矩陣之間具有耦合,引入松弛矩陣進(jìn)行解耦得到便于求解的矩陣形式。所得到的矩陣函數(shù)為無窮維問題,采用網(wǎng)格技術(shù)和近似基函數(shù)法來進(jìn)行矩陣的降維和求解。用本文所設(shè)計的切換控制邏輯完成子系統(tǒng)之間的切換控制。然后,對于給定的參數(shù)變化區(qū)間,設(shè)計一簇子LPV系統(tǒng)輸出反饋控制器,保證其閉環(huán)穩(wěn)定且滿足給定的魯棒H∞性能指標(biāo)。由于所需求解的矩陣函數(shù)都為無窮維的,同樣采用網(wǎng)格技術(shù)和近似基函數(shù)法將求解矩陣轉(zhuǎn)化為參數(shù)矩陣。構(gòu)造控制器時需要求矩陣函數(shù)的導(dǎo)數(shù),但調(diào)度參數(shù)的變化率并非實時可測,所以可以固定所求矩陣中的一個,使控制器矩陣中不含導(dǎo)數(shù)項。用本文所設(shè)計的切換控制邏輯完成子系統(tǒng)之間的切換控制。
[Abstract]:In the field of control, with the rapid development of technology and the continuous improvement of control quality, it is impossible to meet the control needs by relying solely on a certain system or the control mode of a certain system. The hybrid system can solve this problem. The switched system is a typical hybrid system. In short, the switched system is composed of a series of subsystems and the switching logic to coordinate these subsystems. In order to achieve the optimal control effect, the switching control logic of the control system becomes the most important in order to make the operation of the system stable. In recent years, with the wide application of linear parametric variable parameter linear Parameter-Varying (LPV-Varying) system in practical production, the existing problems have emerged. In the design of LPV system, because the parameter range is too large, the designed LPV system can not guarantee the good control effect everywhere in the parameter space. In this paper, a switching control method for linear variable parameter systems is proposed, in which the variable interval of scheduling parameters is divided into several subregions with overlapping regions, and the controllers are designed in these subspaces. The control output of the overlapped part is determined by the two adjacent controllers. The two adjacent subsystems run simultaneously in the overlapped region, and the results are transformed into the system output after the weighted function. When the scheduling parameter changes with time and enters the corresponding sub-interval, the corresponding subsystem is called. At the same time, in order to reduce the conservatism of the designed system, the handover times within the average residence time will also be taken into account. The main work of this paper is as follows: firstly, for a given parameter variation interval, a cluster of LPV state feedback controllers are designed to ensure the closed-loop stability and satisfy the given robust H 鈭，

本文編號：2037247