Search

The partial-nodes-based state estimation (PNBSE) problem is investigated for discrete time-varying nonlinear complex networks subject to stochastic inner coupling strength (SICS) and random sensor delay. The SICS is described by multiplicative noise. The random sensor delay with uncertain occurrence probability is modeled by a sequence of random variables, which is governed by Bernoulli distribution. For the sake of reducing the network transmission burden, a novel state estimator is constructed based on an event-triggered communication mechanism. By means of solving two recursive matrix equations, the upper bound of estimation error covariance (UBEEC) is calculated, and the appropriate gain matrix is selected to minimize the obtained UBEEC. Furthermore, the performance of the proposed estimation algorithm is verified, where the monotonicity analysis is shown between the trace of UBEEC and the occurrence probability of random sensor delay. Conclusively, a simulation example is adopted to illustrate the feasibility of PNBSE method. [ABSTRACT FROM AUTHOR]

In this paper, the finite-time fault detection (FTFD) problem is investigated for a class of discrete-time nonlinear systems subject to the probabilistic interval time-varying delay and the randomly occurring faults (ROFs) under the annulus-event-based communication strategy (AEBCS). A Bernoulli stochastic variable is used to depict the probability interval time-varying delay phenomenon, where the time delay is bounded and its probability distributions can be observed. The phenomena of the ROFs are characterized by the Markov chain with two states. Furthermore, the AEBCS is applied to determine whether the measured signals can be transmitted to the filter or not. For the fault detection (FD) problem, we propose some sufficient conditions to ensure that the error dynamics system is finite-time stochastically stable with the specified H ∞ performance index. Meanwhile, the gains of the filter can be calculated via the feasible solution to certain linear matrix inequalities. In the end, two numerical examples are proposed to verify the effectiveness of newly proposed FTFD method via the AEBCS. [ABSTRACT FROM AUTHOR]

This paper discusses the event-triggered set-membership state estimation problem for discrete delayed linear systems over sensor networks, where noises are unknown but bounded. For the purpose of saving limited communication resources between sensor nodes, a novel adaptive dynamic event-triggered mechanism is applied to reduce the frequency of data transmission. Based on the topological structure of sensor networks and the event-triggered condition, a distributed set-membership state estimator is designed to ensure that the states and estimation errors are restricted within the corresponding zonotopes. The appropriate gain matrices of the estimator are selected to minimize the F-radius of the zonotopes, where the augmented estimation error is located at every instant. In order to avoid the huge computational load caused during the estimation iteration process, the reduction operator is introduced to find a compromise between the conservative of the zonotopic estimation results and the complexity of operation. Finally, a numerical example is offered to demonstrate the effectiveness of the proposed event-triggered zonotopic set-membership state estimation algorithm. [ABSTRACT FROM AUTHOR]

This paper deals with the problem of finite-time stabilization for the multi-rate networked control system with network-induced time delays. In order to guarantee the finite-time stability and overcome the adverse impact of network-induced time delays in both sensor-to-controller channel and controller-to-actuator channel, a networked predictive control strategy is proposed based on multi-rate sampling mechanism. By utilizing the techniques of lifting and augmenting, the augmented closed-loop system is derived with a uniform sampling rate. By applying the Lyapunov stability theorem, sufficient conditions are established such that the multi-rate networked control system can be finite-time stabilized. Furthermore, the gain matrices are determined for the observer and the controller by solving bilinear matrix inequality or linear matrix inequality. Finally, a numerical example is presented to show the validity of the obtained results. [ABSTRACT FROM AUTHOR]

In this paper, the problem of robust fault detection is investigated for nonlinear uncertain Markovian jump systems subject to time-varying delays, Lipschitz nonlinearities and parameter uncertainties under the event-triggered protocol. The event-triggered mechanism is introduced to adjust the transmission frequency of the data sent to the remote module. An event-based fault detection method is presented to guarantee the sensitivity of residual to fault and the attenuation of the effect from the disturbance on the residual. In this way, the considered fault detection problem is solvable by testifying the feasibility of H ∞ filtering problem. By constructing the mode-dependent Lyapunov functional, new sufficient criteria are derived in terms of the linear matrix inequality technique, which insures the stochastic stability with prescribed performance for addressed MJSs. In the end, a numerical example is exploited and the validity of newly presented event-based fault detection method is shown. [ABSTRACT FROM AUTHOR]

This paper is concerned with the robust fault detection (FD) problem for discrete uncertain systems with data drift, randomly occurring faults (ROFs) and randomly occurring nonlinearities (RONs) under weighted try-once-discard (WTOD) protocol. The phenomena of the RONs are characterized by the Bernoulli distributed variable. The Markov chain with two states is introduced to describe the ROFs in the system. The data drift is depicted by an array of mutually independent stochastic variables with individual probabilistic distribution functions. Moreover, the WTOD protocol is utilized to lessen the incidence of data collisions and enhance the communication efficiency. An FD filter is established, and sufficient conditions are proposed under which the resultant filtering error system is stochastically stable and achieves a prescribed H ∞ performance irrespective of the simultaneous presence of the WTOD protocol mechanism, ROFs, RONs and data drift. Besides, the form of the desired filter parameters is presented in terms of the solutions to certain matrix inequalities. Finally, two examples are exploited to demonstrate the feasibility of the obtained robust FD algorithm. [ABSTRACT FROM AUTHOR]

In this paper, the non-fragile distributed fault detection (FD) problem is discussed for delayed Markov jump systems over sensor networks under the weighted try-once-discard protocol, where the randomly switching nonlinearities and the randomly occurring faults are considered. Specifically, a non-fragile distributed FD filter is constructed, and the compact residual dynamic system is obtained in terms of augmentation technique. Subsequently, by applying the Lyapunov stability theory and Wirtinger inequality, sufficient conditions are derived to ensure that the nominal system corresponding to residual dynamic system is stochastically stable and stochastically strictly (Q1,Q2,Q3)\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$(Q_1,Q_2,Q_3)$$\end{document}-γ\documentclass[12pt]{minimal}\usepackage{amsmath}\usepackage{wasysym}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{mathrsfs}\usepackage{upgreek}\setlength{\oddsidemargin}{-69pt}\begin{document}$$\gamma $$\end{document} dissipative. On this basis, the explicit expressions of FD filter matrices are obtained by means of the linear matrix inequality method. At last, a simulation is provided to demonstrate the effectiveness of the proposed distributed FD scheme. [ABSTRACT FROM AUTHOR]