摘要
针对一类具有不确定网络诱导时延和范数有界未知输入影响的网络控制系统(Networked Control System s,简称NCS),研究了故障诊断滤波器的设计问题。通过构造基于观测器的滤波残差系统,将故障诊断滤波器的设计问题归结为H∞滤波问题。通过引入2个附加的松弛矩阵使Lyapunov矩阵和系统矩阵分离,得到了一个新的鲁棒H∞性能判据,并基于该判据设计了系统的鲁棒H∞滤波器。设计的滤波器不但保证闭环系统的二次稳定性,同时使残差系统的H∞范数小于一个给定的衰减水平γ,并且设计了一个迭代算法来得到问题的优化解。最后通过一个仿真算例验证了所提出算法的可行性和有效性。
The fault detection filter design problem of networked control systems with uncertain network-induced delay and norm bounded unknown input is investigated, which is formulated as an H∞filter design problem by constructing residual generator and fault detection filter based on observer. An H∞ performance criterion is presented by introducing two slack matrices to separate the systems matrices and the Lyapunov matrices. An approach to design robust H∞ filter is achieved based on the criterion. The designed controller guarantees the quadratic stability of the closed-loop system and the H∞ norm bound within a given constant. An iterative algorithm is developed to further refine the optimal filter. A numerical example shows the effect of the proposed approach.
出处
《控制工程》
CSCD
北大核心
2009年第3期354-357,362,共5页
Control Engineering of China
基金
国家自然科学基金资助项目(60504008)
关键词
网络控制系统
故障诊断
观测器
参数依赖
networked control system
fault detection
observer
parameter-dependence
