Robust fault detection for uncertain networked control systems with random data packet loss

In this paper, a problem of robust fault detection for uncertain networked control system (NCS) including random data packet loss is investigated. This NCS is established as a stochastic system with norm-bounded uncertainty, in which the stochastic variable is followed by Bernoulli distribution. Then, a robust fault detection filter (FDF) is designed based on the observer method, which is obtained by solving linear matrix inequality(LMI). Besides, the sufficiency of asymptotically mean-square stable is achieved. It is worth noting that a threshold with considering false alarm rate (FAR) is presented in the residual evaluation part. Finally, a numerical example illustrates the effectiveness of the proposed approach.