广义马尔科夫跳变系统的事件触发异步故障检测滤波

Event-Triggered Asynchronous Fault Detection Filtering for Singular Markov Jump Systems

摘要:

针对一类具有事件触发机制的离散时间广义马尔科夫跳变系统，解决了异步故障检测滤波问题。 提出了一种事件触发策略，实现了合理利用网络资源，降低通信负担的目的；考虑了系统模态和滤波器模态之间存在的异步现象，使用两种不同的马尔科夫链来分别表示系统模态和滤波器模态，并在此基础上提出了一种异步故障检测滤波器的设计方法。 通过将异步故障检测滤波器的设计问题与事件触发机制相结合，根据Lyapunov稳定性理论，给出了保证残差误差系统具有正则性、因果性和随机稳定的并且满足 性能的充分条件，通过求解一组特定的线性矩阵不等式，得到了期望的异步故障检测滤波器参数。 最后，通过一个仿真例子说明了所设计的异步故障检测滤波器的设计方法是有效的。

Abstract:

For a class of discrete-time singular Markov jump systems with an event-triggered mechanism, the problem of asynchronous fault detection filtering is solved. First, an event-triggered strategy is proposed to achieve the purpose of rationally using network resources and reducing the burden of communication. Then, considering the asynchronous phenomenon between the system mode and the filter mode, two different Markov chains are used to represent the system mode and the filter mode respectively, and on this basis, a design method of the asynchronous fault detection filter (FDF) is proposed. Next, by combining the design problem of the asynchronous fault detection filter with the event-triggered mechanism, based on the Lyapunov stability theory, the sufficient condition is given to ensure that the residual error system is to be regular, causal, stochastically stable and satisfy performance. The expected parameters of the asynchronous fault detection filter are available by solving a specific set of linear matrix inequality (LMI). Finally, a simulation example is given to illustrate that the design method of the asynchronous fault detection filter is effective.