CASIA OpenIR  > 毕业生  > 博士学位论文
基于观测器的网络化控制系统故障检测
Alternative TitleObserver Based Fault Detection For Networked Control Systems
陈煜强
Subtype工学博士
Thesis Advisor刘国平
2006-06-30
Degree Grantor中国科学院研究生院
Place of Conferral中国科学院自动化研究所
Degree Discipline控制理论与控制工程
Keyword网络化控制系统 网络延时 故障检测 Netcon 组态软件 Networked Control System Network Delay Fault Detection Netcon Configuration Software
Abstract网络化控制系统是通过通信网络构成反馈回路的控制系统,它是伴随着自动化技术、网络技术及计算机技术的发展而产生并发展起来的,随着基于以太网的网络化控制系统成为控制界发展和研究的热点,也促进了这样的网络化控制系统故障诊断技术的发展。由于共享网络介质给控制系统带来的一系列问题,如网络延时、多包传输及丢包等问题,影响了系统的控制性能,也影响了对网络化控制系统的故障诊断。因此,研究基于以太网的网络化控制系统的故障检测具有非常重要的理论意义和实用价值。 论文的主要内容和创新之处如下: 1)、考虑通信网络对系统的影响,利用NetCon系统智能节点的强大计算处理能力,把故障检测 模块与控制器模块配置在一起,故障检测模块能够得到控制器的输出,这样在进行故障检测时可以仅考虑反馈通道的网络延时,而不考虑前馈通道的网络延时;并且避免了闭环控制对故障检测的影响,实现了对系统的开环故障检测。针对系统运行过程中,模型出现故障的情况下,采用基于观测器的方法并结合在线辨识技术,对于传感器的滞后输出采用自适应预报技术作为补偿方式,基于观测器的故障检测算法作为在线监控模块,实现了对NCS的在线故障检测,并通过NetCon系统组态软件进行实时的监控。本文通过离线的仿真和在线的实际电机实验验证了该方案对NCS进行故障检测的有效性。 2)、根据NCS中数据传输延时的随机性特点,本文采用了Markov延时模型,对于网络延时服从离散Markov链的情况,研究了NCS的故障检测问题。对于附有时间戳的数据包,故障检测模块在每个模式下,可以在线获得网络延时的大小,从而,在该模式下可看作是固定时延系统,随机性表现为了随机切换的特点,文章分析了这样的系统的稳定性;当系统有模型误差等扰动 时,给出了 稳定性的充分条件,并设计了在有随机网络延时情况下,基于LMI的NCS的故障检测算法。最后分别对无扰动和有扰动两种情况,当传感器输出延时服从Markov链的情况,进行数字仿真来验证提出的故障检测方案。 3)、根据可视化组态软件的需求,考虑到软件的通用性和可扩展性,按照软件工程的理论使用UML工具进行设计,开发了NetCon系统中的主要组成部分——NetCon组态软件,详细介绍了软件的结构设计、图形界面的设计和虚拟仪器的设计和实现过程,并为NetCon系统提供了完整的图形组态功能,用户在ICDraw集成图形开发环境中可任意设计图形监控模块。 4)、对报警、事件模块进行了分析与设计,报警系统作为监控组态软件的重要组成部分,是软件价值的重要体现方面;报警事件模块处于软件数据库核心区。文章根据报警事件模块的需求,按照组件化设计的思想和三层软件模型结构进行设计,分别是:数据服务层、业务逻辑层和表示层,最终设计为独立的组件,可以与其它组件结合组成分布式应用系统。 论文最后对所取得的研究成果进行了总结。
Other AbstractWith the development of network technology, more and more networks (e.g. Internet) have been applied to distributed control systems, whose feedback loop is closed through a real-time network, which is termed the networked control system (NCS). With the development of NCS, the fault diagnosis of NCS has become a hot subject for the security of the NCS. With the introducing of networks in control systems, there exist many problems due to sharing a common network medium, such as network-induced delay, data packet dropout and multi-packets transmission. So, research on fault detection of networked control systems based on Ethernet is of great significance in both theory and practice. Main work and contributions in this dissertation are described as follows: 1). With the strong computation power of the NetCon intelligent node, the fault detection module and control module are placed at the same site. With this configuration, the effect of the closed-loop control for the fault detection and the forward delay do not need to be considered. When a fault happens in the model, the adaptive prediction technique is applied for compensating for sensor delays. Using the observer-based fault diagnosis module and the NetCon configuration software, the on-line fault detection and monitoring for the NCS is achieved. An off-line simulation and a motor control experiment are used to test the feasibility of the proposed method. 2). According to the stochastic properties of the data transmission, the Markov delay model is applied to study the fault detection problem of NCS. With the time stamp in the data packets, at the every mode, the delay value can be obtained and the constant time delay theory can be applied. The randomness of the delay denotes stochastic switching. The dissertation gives the stabililty analysis and sufficient conditions for control. A LMI-based fault detection algorithm for the NCS is proposed. At last, the simulation experiment is used to test the feasibility. 3). According to the demands of the graph interface of the visual supervisory configuration software, the dissertation details the design process of some functions of the software, which considers its generalization and expansion with the tool of the UML. The user can design any graph monitoring module freely in this software environment. The NetCon system provides many powerful configuration functions,such as, figure configuration and virtual instrument. 4). The alarm and event modules are the important part of the NetCon system. It embodies the software value. Following the requirements of the NetCon system, the dissertation gives the analysis and design of the alarm and event module. The component can be combined into a distributed system with other components. The conclusions are given in the final chapter.
shelfnumXWLW1051
Other Identifier200318014602968
Language中文
Document Type学位论文
Identifierhttp://ir.ia.ac.cn/handle/173211/5952
Collection毕业生_博士学位论文
Recommended Citation
GB/T 7714
陈煜强. 基于观测器的网络化控制系统故障检测[D]. 中国科学院自动化研究所. 中国科学院研究生院,2006.
Files in This Item:
File Name/Size DocType Version Access License
CASIA_20031801460296(6588KB) 暂不开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[陈煜强]'s Articles
Baidu academic
Similar articles in Baidu academic
[陈煜强]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[陈煜强]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.