Fault Estimation For Continuous-Time Non-Linear Switched Systems With Time-Varying Delay Based On Intermediate Estimator

In this work, the fault estimation (FE) problem is studied for continuous-time non-linear switched systems with time-varying delay over fading channel under average dwell time switching. A novel FE method based on the intermediate estimator is used to eliminate the observer matching constraint condition, where an intermediate variable is introduced to estimate the state and fault simultaneously. Meanwhile, the fading channel of data transmission is modelled as Bernoulli distributed white sequences. Furthermore, because the observer matching conditions are eliminated by the intermediate estimator, both the wirtinger integral inequality and an advanced reciprocal convex combination method are utilised to provide a tighter estimation on continuous derivative of certain Lyapunov-Krasovskii function over fading channel, leading to a less conservative result. Moreover, the solution of the parameters for intermediate estimator is characterised in the feasibility of linear matrix inequality problem. Finally, two numerical examples indicate the proposed technique is valid and superior.