具有通信受限和丢包的Delta算子网络系统故障检测研究

发布时间：2018-05-11 11:28

  本文选题：网络控制系统 + 通信受限　； 参考：《郑州大学》2017年硕士论文

【摘要】：随着通信技术的进步和控制系统规模的扩大,数字网络被用来传输控制系统各节点间信息,网络控制系统便由此诞生。与点对点传输模式的控制系统相比,网络控制系统的布线较少、安装简单、信息共享并且系统的扩展和维护比较方便。然而,由于带宽的限制,作为传输信息媒介的网络会带来时延、通信受限、数据包丢失等问题,这些因素必然会影响系统性能,甚至会造成系统不稳定。网络控制系统在生活的各个方面都有应用,故障的出现会给人们带来极大的影响,及时检测出系统的故障并报警显得尤为重要。由于网络控制系统中采用数字网络作为节点间信息的传输媒介,在相应的研究和设计中需要将连续系统进行离散化。现有文献大多数使用移位算子的方法对系统进行离散化,而在高速采样的情况下,利用移位算子离散得到的系统并不趋近于离散前的连续系统,并且其极点不完全在稳定环内,使用Delta算子离散化方法则能避免这种缺点。采用Delta算子方法结合网络带来的问题对网络控制系统进行分析和设计,是目前的研究热点和重点。本文是作者对参加的河南省教育厅技术研究重点资助项目《网络环境下基于Delta算子的故障检测与滤波方法》(项目编号:14A120004),和国家自然科学基金资助项目《高速采样网络系统故障检测与滤波的统一方法》(项目编号:61471323)所做研究的总结。本文采用Delta离散化方法研究了通信受限和丢包环境下网络系统故障检测的问题,创新点和研究成果如下:(1)针对具有通信受限和数据包丢失的Delta算子系统,设计故障检测观测器。采用通信序列的方法解决通信约束,使用马尔科夫链描述系统的丢包,将所研究的系统建模为马尔科夫切换系统。利用随机稳定性理论和线性矩阵不等式(LMIs)工具箱,得到系统具有随机稳定性和H_∞性能的故障观测器的增益值。(2)在上一研究的基础上,把系统的时延考虑在内,建立通信受限和丢包现象同时存在于传感器到控制器、控制器到执行器链路的时延Delta算子模型。设计滤波器进行参数估计,从而产生残差系统。使用Delta域的Lyapunov泛函数技术和线性矩阵不等式(LMIs)工具箱,得到系统满足渐近稳定性和H_∞性能的充分条件。并用凸优化方法求其相应的解,对算法的可行性进行仿真验证。(3)研究了通讯约束和丢包情况下时延网络控制系统鲁棒故障检测的问题。将系统建模为参数不确定的Delta算子模型,使用状态反馈控制律方法和MultiLyapunov函数技术设计满足渐近稳定和H_∞性能的故障检测滤波器。并在现有的评价函数和阈值的基础上提出新的故障检测的标准。通过仿真对比,所设计的算法对故障检测有鲁棒性和更好的实时性。
[Abstract]:With the development of communication technology and the expansion of control system scale, digital network is used to transmit information between nodes of control system, and network control system is born. Compared with the control system with point-to-point transmission mode, the network control system has less wiring, simple installation, information sharing and easy system expansion and maintenance. However, due to the limitation of bandwidth, the network as a transmission information medium will bring delay, limited communication, packet loss and other problems. These factors will inevitably affect the performance of the system, and even lead to system instability. Network control system has been applied in every aspect of life. The emergence of fault will bring great influence to people. It is very important to detect the fault and alarm in time. Because the digital network is used as the transmission medium of the information between nodes in the networked control system, it is necessary to discretize the continuous system in the corresponding research and design. In most references, the shift operator is used to discretize the system. In the case of high speed sampling, the discrete system obtained by shift operator does not approach the continuous system before discretization, and its poles are not completely in the stable loop. This shortcoming can be avoided by using Delta operator discretization method. Using Delta operator method to analyze and design the networked control system combined with the problems brought by the network is the focus of research at present. This paper is about the key technical research project of Henan Provincial Education Department < Fault Detection and filtering method based on Delta operator in Network Environment "(Project No.: 14A120004, and National Natural Science Foundation of China < High-Speed sampling)" "the unified method of fault detection and filtering in network system." Summary of research done by item No.: 61471323. In this paper, Delta discretization method is used to study the problem of network system fault detection in communication constrained and packet loss environments. The innovations and research results are as follows: 1) for Delta operator systems with communication limitation and packet loss, a fault detection observer is designed. The method of communication sequence is used to solve the communication constraint, and the Markov chain is used to describe the packet loss of the system, and the studied system is modeled as Markov switching system. By using the stochastic stability theory and the linear matrix inequality (LMI) toolbox, the gain value of the fault observer with stochastic stability and H _ 鈭，

本文编号：1873760