Detection of intermittent faults for nonuniformly sampled multi-rate systems with dynamic quantisation and missing measurements

ABSTRACT

This paper is concerned with the fault detection problem for a class of networked multi-rate systems with nonuniform sampling and dynamic quantization. The sampling interval of the measurements is allowed to be nonuniform that is governed by a time-homogenous Markov process with partly unknown and uncertain transition probabilities. The measured output is quantized by a dynamic quantizer and then transmitted through communication network subject to data missing. The main purpose of the problem under consideration is to design sampling-interval-dependent fault detection filters such that, in the simultaneous presence of nonuniform sampling, dynamic quantization, intermittent faults as well as missing measurements, the robustness of residuals with respect to the disturbance and the sensitivity of the residuals against the fault are guaranteed. Finally, a three-tank system is utilized to illustrate the effectiveness of the proposed fault detection scheme.     

Convergence Analysis of the Hierarchical Least Squares Algorithm for Bilinear-in-Parameter Systems

This paper studies the convergence of the hierarchical identification algorithm for bilinear-in-parameter systems. By replacing the unknown variables in the information vector with their estimates, a hierarchical least squares algorithm is derived based on the model decomposition. The proposed algorithm has higher computational efficiency than the over-parameterization model-based recursive least squares algorithm. The performance analysis shows that the parameter estimation errors converge to zero under persistent excitation conditions. The effectiveness of the proposed algorithm is verified by simulation examples.     

Stochastic Stability for a Class of Discrete-time Switched Neural Networks with Stochastic Noise and Time-varying Mixed Delays

In this paper, stochastic stability is analyzed for a class of discrete-time switched neural networks, in which time-varying mixed delays and stochastic noise are considered. Specifically, benefitting from the triple summation term included in a new Lyapunov functional, time-varying distributed delays are tackled and a criterion of decay estimation for a non-switched neural network is firstly obtained. Subsequently, in view of average dwell time methodology and stochastic analysis, several sufficient conditions are obtained to ensure that the stochastic stability problem is solvable. Furthermore, the derived sufficient conditions reflect that the decay rate of the considered neural networks has a close relationship with average dwell time, upper and lower bounds of delays and intensity of stochastic noise. Finally, validity of the inferred conclusions is given by a simulated example.     

Security‐guaranteed filtering for discrete‐time stochastic delayed systems with randomly occurring sensor saturations and deception attacks

In this paper, the security‐guaranteed filtering problem is studied for a class of nonlinear stochastic discrete time‐delay systems with randomly occurring sensor saturations (ROSSs) and randomly occurring deception attacks (RODAs). The nonlinearities in systems satisfy the sector‐bounded conditions, and the time‐varying delays are unknown with given lower and upper bounds. A novel measurement output model is proposed to reflect both the ROSSs and the RODAs. A new definition is put forward on the security level with respect to the noise intensity, the energy bound of the false signals, the energy of the initial system state, and the desired security degree. We aim at designing a filter such that, in the presence of ROSSs and RODAs, the filtering error dynamics achieves the prescribed level of security. By using the stochastic analysis techniques, a sufficient condition is first derived under which the filtering error system is guaranteed to have the desired security level, and then, the filter gain is designed by solving a linear matrix inequality with nonlinear constraints. Finally, a numerical example is provided to demonstrate the feasibility of the proposed filtering scheme. Copyright © 2016 John Wiley & Sons, Ltd.     

Distributed optimisation for multi-agent systems with the first-order integrals under Markovian switching topologies

This paper studies the distributed optimisation problem for multi-agent systems with the first-order dynamics over Markovian switching topologies. The interaction topology among agents’ switches following a Markov process and each topology is modelled as a state of the Markov process. The aim is to minimise the global cost functions and make the agents converge to the optimal point through the network communication between the agents, where each agent has a local convex cost function only known by itself. Utilising the knowledge of convex analysis and graph theory, we establish a distributed algorithm for the optimisation problem with randomly switching topologies. A sufficient condition for the existence of such algorithm is obtained by using the Lyapunov method. Besides, the result is also extended to the cases of a Markov process with partially unknown transition rates. Finally, numerical simulations are given to validate the proposed algorithm.     

Sampled‐data consensus of nonlinear multiagent systems subject to cyber attacks

This paper is concerned with the sampled‐data consensus problem for a class of nonlinear multiagent systems subject to cyber attacks. The considered attacks are assumed to be recoverable, which destroy the connectivity of the network topology with directed spanning tree. In light of the designed sampled‐data control protocol, a more general switched system is proposed to model both the cyber attacks and the sampled‐data mechanism within a unified framework. Then, by using the contradiction methods, the solution of a class of differential equations is provided to deal with the technical challenge resulting from the switching and sampling issues. Furthermore, in terms of such a solution combined with the constructed piecewise Lyapunov function, sufficient conditions are derived under which the nonlinear multiagent systems subject to attacks achieve the consensus exponentially. The relationship between the attack frequency and the sampling period is also revealed through the obtained conditions. Finally, simulations are given to show the effectiveness of the proposed results.     

Dynamic self‐triggered output‐feedback control for nonlinear stochastic systems with time delays

In this paper, the dynamic self‐triggered output‐feedback control problem is investigated for a class of nonlinear stochastic systems with time delays. To reduce the network resource consumption, the dynamic event‐triggered mechanism is implemented in the sensor‐to‐controller channel. Criteria are first established for the closed‐loop system to be stochastically input‐to‐state stable under the event‐triggered mechanism. Furthermore, sufficient conditions are given under which the closed‐loop system with dynamic event‐triggered mechanism is almost surely stable, and the output‐feedback controller as well as the dynamic event‐triggered mechanism are co‐designed. Moreover, a dynamic self‐triggered mechanism is proposed such that the nonlinear stochastic system with the designed output‐feedback controller is stochastically input‐to‐state stable and the Zeno phenomenon is excluded. Finally, a numerical example is provided to illustrate the effectiveness of proposed dynamic self‐triggered output‐feedback control scheme.     

Adaptive fuzzy output‐feedback tracking control for switched stochastic pure‐feedback nonlinear systems

This paper considers the problem of adaptive fuzzy output‐feedback tracking control for a class of switched stochastic nonlinear systems in pure‐feedback form. Unknown nonlinear functions and unmeasurable states are taken into account. Fuzzy logic systems are used to approximate the unknown nonlinear functions, and a fuzzy observer is designed to estimate the immeasurable states. Based on these methods, an adaptive fuzzy output‐feedback control scheme is developed by combining the backstepping recursive design technique and the common Lyapunov function approach. It is shown that all the signals in the closed‐loop system are semiglobally uniformly ultimately bounded in mean square in the sense of probability, and the observer errors and tracking errors can be regulated to a small neighborhood of the origin by choosing appropriate parameters. Finally, a simulation result is provided to show the effectiveness of the proposed control method.     

Pinning synchronization of fractional‐order complex networks with adaptive coupling weights

This work studies the issue of synchronization control for a type of fractional‐order complex networks, in which the adaptive coupling matrix is considered under the directed topology structure. A pinning control strategy, with the free selection of pinning nodes, is adopted for the synchronization goal. Then, by absorbing the information of eigenvectors and adaptive laws for the coupling matrix, a new Lyapunov function is constructed, by which, and with the assistance of Gronwall inequality and network features, the sufficient condition for Mittag‐Leffler synchronization of the fractional‐order network is established. Accordingly, an easy verifiable algebraic criterion is further derived by means of some matrix inequalities. Besides, we also discuss the effect of outer coupling strength on the achievement of network synchronization. Finally, a numerical experiment is performed to show the evidence of the correctness and effectiveness of the proposed results.