本文選題：分群一致　切入點：Lagrange網(wǎng)絡　出處：《上海大學》2016年博士論文　論文類型：學位論文

【摘要】：近年來具有Lagrange個體動力學結(jié)構(gòu)和特征的網(wǎng)絡化動力學系統(tǒng)協(xié)調(diào)控制己引起人們的廣泛關注：這主要是由于它能描述在復雜集成化生產(chǎn)進程中許多物理和力學對象,其中具有靈活性、機動性、可靠性和可操作性是可期待甚至是必要的特征.另一方面,與完全同步或者一致相比,有些情況下分群一致能更好地處理復雜多智能體系統(tǒng)協(xié)調(diào)控制問題.本文研究了Lagrange網(wǎng)絡分群一致(同步)控制及相關問題,主要工作有以下幾點：一.不確定Lagrange網(wǎng)絡的自適應分群一致.在無循環(huán)劃分和平衡耦合劃分兩種有向分群拓撲結(jié)構(gòu)下,給出了參數(shù)不確定Lagrange網(wǎng)絡的分布式自適應分群一致控制策略.基于幾何圖論和矩陣理論,提出了一種新的分解方法,并在此方法基礎上討論了受控Lagrange系統(tǒng)的穩(wěn)定性,闡述了Lagrange網(wǎng)絡實現(xiàn)分群一致的充分必要條件.所給的控制算法有以下兩個特點：在無循環(huán)劃分網(wǎng)絡下,網(wǎng)絡的無循環(huán)結(jié)構(gòu)就能夠保證實現(xiàn)分群一致；而在平衡耦合劃分拓撲下,簡單的代數(shù)準則能夠?qū)崿F(xiàn)系統(tǒng)的分群目標.進一步地,當網(wǎng)絡拓撲是有向無循環(huán)劃分網(wǎng)絡時,通過設計一種積分控制器,系統(tǒng)的分群一致狀態(tài)能夠顯式地表達出來,因而提供了一種實現(xiàn)預定分群一致目標的系統(tǒng)的控制方法.二.受擾動的不確定Lagrange網(wǎng)絡的區(qū)域分群一致.在無循環(huán)劃分分群網(wǎng)絡結(jié)構(gòu)下,研究了網(wǎng)絡化參數(shù)不確定Lagrange系統(tǒng)自適應分群區(qū)域一致控制問題,并對沒有引導者和具有引導者這兩種情況分別提出了相應的自適應區(qū)域分群一致控制算法.對于任意預先給定的一致誤差界,本文給出的充分條件總是能夠保證達到期望的區(qū)域分群一致.與已有的結(jié)果比較,所給的控制算法能夠?qū)崿F(xiàn)全局的穩(wěn)定性,即對任意的有限初值狀態(tài),所給的算法都能夠?qū)崿F(xiàn)系統(tǒng)的區(qū)域分群一致.三.基于動力學基本方程的Lagrange網(wǎng)絡的分群同步.利用動力學基本方程理論,從分析動力學的角度提出了一種網(wǎng)絡化Lagrange系統(tǒng)的優(yōu)化控制方法框架.提出的控制算法的主要特點是在控制要求(約束)中引入了網(wǎng)絡的拓撲結(jié)構(gòu).所給的算法在結(jié)構(gòu)上可以分成兩部分：第一部分描述了網(wǎng)絡的結(jié)構(gòu),第二部分是系統(tǒng)的主動反饋控制.基于圖論和矩陣理論,給出了網(wǎng)絡化Lagrange系統(tǒng)實現(xiàn)分群同步的充分條件.最后,由陀螺儀構(gòu)成的網(wǎng)絡系統(tǒng)作為數(shù)值模擬對象來說明了這種優(yōu)化控制算法中各個參數(shù)的作用,并驗證了所給理論結(jié)果的有效性.
[Abstract]:In recent years, the coordinated control of networked dynamic systems with individual dynamics structure and characteristics of Lagrange has attracted wide attention: this is mainly because it can describe many physical and mechanical objects in complex integrated production processes. Flexibility, mobility, reliability and maneuverability are desirable and even necessary features. On the other hand, compared to complete synchronization or consistency, In some cases, clustering uniformity can better deal with the coordinated control problem of complex multi-agent systems. In this paper, cluster consistent control and related problems in Lagrange networks are studied. The main works are as follows: 1. The adaptive clustering of uncertain Lagrange networks is consistent. A distributed adaptive cluster consistent control strategy for Lagrange networks with uncertain parameters is presented. Based on geometric graph theory and matrix theory, a new decomposition method is proposed, and the stability of controlled Lagrange systems is discussed. In this paper, the necessary and sufficient conditions for clustering uniformity in Lagrange networks are described. The proposed control algorithm has the following two characteristics: in the case of non-cyclic partition networks, the no-cyclic structure of the network can guarantee the clustering uniformity, while in the balanced coupling partition topology, Moreover, when the network topology is directed uncircularly partitioned network, by designing an integral controller, the cluster uniform state of the system can be expressed explicitly. Therefore, this paper provides a control method for the system to achieve the consistent target of predetermined clustering. Secondly, the region clustering of disturbed uncertain Lagrange networks is consistent. Under the network structure of non-cyclic partitioning clustering, In this paper, the problem of adaptive cluster region consistent control for networked parameter uncertain Lagrange systems is studied. The corresponding adaptive region clustering consistent control algorithm is proposed for the two cases of no guide and one with guide. For any given consistent error bound, The sufficient conditions given in this paper can always ensure that the desired regional clustering is consistent. Compared with the existing results, the proposed control algorithm can achieve global stability, that is, for arbitrary finite initial value states, The proposed algorithms can realize the regional cluster consistency of the system. Thirdly, the cluster synchronization of Lagrange networks based on the fundamental equations of dynamics is used. An optimal control method framework for networked Lagrange systems is proposed from the point of view of analysis dynamics. The main feature of the proposed control algorithm is the introduction of the network topology into the control requirements (constraints). The structure can be divided into two parts: the first part describes the structure of the network, The second part is the active feedback control of the system. Based on graph theory and matrix theory, the sufficient conditions for the realization of cluster synchronization in networked Lagrange systems are given. The network system composed of gyroscopes is used as a numerical simulation object to illustrate the function of the parameters in the optimal control algorithm and verify the validity of the theoretical results given.
【學位授予單位】：上海大學
【學位級別】：博士
【學位授予年份】：2016
【分類號】：O231

【參考文獻】

相關期刊論文 前5條

1 ZHOU Jin;WU XingJie;LIU ZengRong;;Distributed coordinated adaptive tracking in networked redundant robotic systems with a dynamic leader[J];Science China(Technological Sciences);2014年05期

2 WANG Qiang;WANG YuZhen;;Cluster synchronization of a class of multi-agent systems with a bipartite graph topology[J];Science China(Information Sciences);2014年01期

3 ZHAO Han;ZHEN Shengchao;CHEN Ye-Hwa;;Dynamic Modeling and Simulation of Multi-body Systems Using the Udwadia-Kalaba Theory[J];Chinese Journal of Mechanical Engineering;2013年05期

4 林泰明;李東旭;陳浩;;基于約束運動理論的航天員艙外救生中的姿態(tài)運動控制[J];宇航學報;2010年02期

5 徐鑒;裴利軍;;時滯系統(tǒng)動力學近期研究進展與展望[J];力學進展;2006年01期

，

本文編號：1634788