An iterative approach to H₋/H_∞ Fault Detection Observer Design for Discrete‐Time Uncertain Systems

Abstract

In this paper, H−/H∞ multi‐objective fault detection observer design problem for linear discrete‐time systems is studied. Owing to the model inaccuracy in practical situations, system uncertainty is considered as well as the structure of the uncertainty. Sufficient conditions for the existence of the observers to guarantee the fault sensitivity and disturbance robustness in infinite frequency domain are presented. A new iterative linear matrix inequality (LMI) algorithm for the observers are proposed to obtain the solutions. In order to reduce the conservativeness, the observer design problem in finite frequency domain is also investigated. Simulation results illustrate the effectiveness of the proposed methods.

H_/H∞ fault detection observer design for a polytopic LPV system using the relative degree

“…In the simulations, the initial state is chosen as [0 0 0 0] T , the input vector is given as u(t) = [sin(0.01t) cos(0.05t + π/2)] T , the system uncertainty w(t) and measurement noise v(t) are set as random signals bounded by [−1, 1]. In this paper, residual evaluation is considered to be the root mean square (RMS) value of the generated residual as in the work of Huang et al (2017). In this paper, two fault scenarios are assumed to be detected.…”

Distributed H∞ Consensus Fault Detection for Uncertain T‐S Fuzzy Systems with Time‐Varying Delays Over Lossy Sensor Networks

“…Consequently, a fault detection technology is needed to monitor the system behaviour by indicating the occurrence, location and size of the fault. Over the past decades, fruitful results regarding fault detection have been obtained [10][11][12][13][14][15][16][17][18][19]. However, most of these results on fault detection focus on the centralised architecture, which may not be suitable to distributed sensor networks.…”

“…One such challenge is the network-induced packet-dropout phenomenon that has recently drawn considerable research attention. Various approaches have been presented in the literature to model the packet dropout phenomenon [6][7][8][9], [13][14][15][16][17]. It should be pointed out that, almost all the literature concerning packet-dropout phenomenon have been specifically considered for networked control systems (NCSs) [16,17,25].…”

A New Residual Evaluation Function Based Fault Diagnosis for a Kind of Nonlinear Systems

“…In , it is concluded that H ∞ / H ∞ and H − / H ∞ are just special cases of the H i / H ∞ optimization problems, and all share with the unified solution. Furthermore, when switched systems, Markovian jump systems, or measurement packet dropouts, disturbance and fault in finite frequency domains are considered, researches about FDF design in the H ∞ / H ∞ or/and H − / H ∞ optimization framework have been found in , to name a few. In contrast to the linear case, works about optimal FD for nonlinear systems are relatively limited.…”