复杂网络容错同步控制及其应用

发布时间：2018-08-20 11:05

【摘要】：复杂网络在自然界中和现代社会中无处不在,很多现实生活中的系统都可以以复杂网络的模型建模,在过去的几十年里,在理学和工程学的各个领域,复杂网络的同步控制问题掀起了研究热潮。而故障的发生则可能导致系统性能下降,不稳定,甚至灾难性事故。因此,能够使系统在故障发生时仍能保持稳定,正常运行,完成既定目标的容错控制引起了广泛的关注。复杂网络结构复杂,对系统的可靠性要求很高,复杂网络的容错同步控制研究具有重大的意义。然而,迄今为止,复杂网络的容错同步控制仍是一个全新的领域,尚未有其他研究人员发表相关研究成果。本课题来源于国家自然科学基金项目,分别针对非线性系统,也就是,耦合不可重构的复杂网络,和非线性切换系统,也就是,耦合可重构的复杂网络,进行容错同步控制的研究。首先,这篇论文介绍了复杂网络和容错控制的研究现状,包括与复杂网络相似的其他多个体组成的网络,如神经网络等的容错控制研究成果。文中着重介绍了像复杂网络系统这样,由多个体组成的系统,所应用的容错控制思想,和容错控制方法。基于前人的研究成果,提出了相关容错控制方法策略,本文的主要研究成果如下:(1)针对一类线性耦合,耦合不可重构的复杂网络,提出了故障估计和故障调节一体化策略。首先设计了一系列分布式自适应观测器,用以同时获得系统状态和系统故障的信息。然后,基于自适应观测器获得的信息,设计了分散式故障调节方案,用以镇定整个复杂网络系统。最后,由多个单连杆机械臂组成的复杂网络模型做出仿真,验证了所提出的容错控制方法的有效性。(2)针对一类耦合可重构的复杂网络(耦合拓扑结构时变),设计了容错协同控制策略。首先,文中基于领域规则,设计了线性协同控制律,并且提出了复杂网络实现同步的充分条件。然后,文中提出了协同容错控制思想和相应的协同容错控制方法,使复杂网络系统在故障存在的情况下,仍然可以完成同步控制。最后,由多个单连杆机械臂组成的复杂网络模型做出仿真,证明了文中所提出的协同容错控制方法的有效性。
[Abstract]:Complex networks are ubiquitous in nature and modern society. Many real-life systems can be modeled by complex networks. In the past few decades, synchronization control of complex networks has been a hot topic in various fields of science and engineering. Therefore, fault-tolerant control, which can keep the system stable, run normally and accomplish the set goal when the fault occurs, has attracted wide attention. The complex network structure is complex and the reliability of the system is very high. The research on fault-tolerant synchronization control of complex network is of great significance. So far, the fault-tolerant synchronization control of complex networks is still a new field, and no other researchers have published relevant research results. Firstly, this paper introduces the research status of complex networks and fault-tolerant control, including fault-tolerant control of networks composed of many other bodies similar to complex networks, such as neural networks. Based on the previous research results, the relevant fault-tolerant control strategies are proposed. The main research results of this paper are as follows: (1) For a class of linear coupled, coupled and non-reconfigurable complex networks, an integrated strategy of fault estimation and fault regulation is proposed. Firstly, a series of distributed control strategies are designed. Then, based on the information obtained by the adaptive observer, a decentralized fault regulation scheme is designed to stabilize the whole complex network system. Finally, a complex network model composed of multiple single-link manipulators is simulated to verify the proposed fault-tolerant control. (2) A fault-tolerant cooperative control strategy is designed for a class of coupled reconfigurable complex networks (with time-varying topology). Firstly, based on domain rules, a linear cooperative control law is designed, and sufficient conditions for synchronization of complex networks are proposed. Then, the idea and phase of cooperative fault-tolerant control are proposed. The corresponding cooperative fault-tolerant control method enables the complex network system to complete synchronous control even when the fault exists. Finally, the simulation of a complex network model composed of multiple single-link manipulators proves the effectiveness of the proposed cooperative fault-tolerant control method.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：O157.5;TP273

【参考文献】