H-/H∞ memory fault detection filtering design for uncertain systems with finite frequency specifications

This article is concerned with the H-/H-infinity memory fault detection filtering design in finite frequency domain for discrete-time systems with polytopic uncertainties. Under the assumption that both disturbances and faults are restricted to finite frequency ranges, that is, the low, middle, or high frequency range, an admissible memory fault detection filter is developed by using historical information of the system. It is shown that the asymptotic stability with prescribed H-/H-infinity performance of the closed-loop filtering error system is guaranteed. It is also shown that by using historical system outputs along with parameter-dependent Lyapunov functions, the better performance of the resulting fault detection filter is obtained. With the aid of Projection lemma, sufficient conditions for the fault detection filtering design are established by solving an optimization problem in the form of a set of linear matrix inequalities (LMIs). Finally, three examples are presented to demonstrate the effectiveness and advantages of the proposed approach.

Automated summary

This article focuses on the design of H-/H-infinity memory fault detection filters in the finite frequency domain for discrete-time systems with polytopic uncertainties. By using historical information and parameter-dependent Lyapunov functions, better performance of the fault detection filter is obtained. Sufficient conditions for the fault detection filtering design are established by solving an optimization problem in the form of a set of linear matrix inequalities (LMIs), showing the effectiveness and advantages of the proposed approach through three examples.