Distributed fault detection over sensor networks with Markovian switching topologies

This paper deals with the distributed fault detection for discrete-time Markov jump linear systems over sensor networks with Markovian switching topologies. The sensors are scatteredly deployed in the sensor field and the fault detectors are physically distributed via a communication network. The system dynamics changes and sensing topology variations are modeled by a discrete-time Markov chain with incomplete mode transition probabilities. Each of these sensor nodes firstly collects measurement outputs from its all underlying neighboring nodes, processes these data in accordance with the Markovian switching topologies, and then transmits the processed data to the remote fault detector node. Network-induced delays and accumulated data packet dropouts are incorporated in the data transmission between the sensor nodes and the distributed fault detector nodes through the communication network. To generate localized residual signals, mode-independent distributed fault detection filters are proposed. By means of the stochastic Lyapunov functional approach, the residual system performance analysis is carried out such that the overall residual system is stochastically stable and the error between each residual signal and the fault signal is made as small as possible. Furthermore, a sufficient condition on the existence of the mode-independent distributed fault detection filters is derived in the simultaneous presence of incomplete mode transition probabilities, Markovian switching topologies, network-induced delays, and accumulated data packed dropouts. Finally, a stirred-tank reactor system is given to show the effectiveness of the developed theoretical results.     

Fault detection filter design for stochastic time-delay systems with sensor faults

This article considers the fault detection (FD) problem for a class of Itô-type stochastic time-delay systems subject to external disturbances and sensor faults. The main objective is to design a fault detection filter (FDF) such that it has prescribed levels of disturbance attenuation and fault sensitivity. Sufficient conditions for guaranteeing these levels are formulated in terms of linear matrix inequalities (LMIs), and the corresponding fault detection filter design is cast into a convex optimisation problem which can be efficiently handled by using standard numerical algorithms. In order to reduce the conservatism of filter design with mixed objectives, multi-Lyapunov functions approach is used via Projection Lemma. In addition, it is shown that our results not only include some previous conditions characterising H _∞ performance and H _? performance defined for linear time-invariant (LTI) systems as special cases but also improve these conditions. Finally, two examples are employed to illustrate the effectiveness of the proposed design scheme.     

Simultaneous reduced-order control and fault detection for discrete-time switched systems with relaxed persistent dwell time regularity

This paper focuses on the problem of simultaneous control and fault detection (FD) for discrete-time switched systems. The main goal is to develop a control/detection unit (CDU) associated with a switching law to control the system and detect faults simultaneously. When the switched systems are with partial measurable states, we directly use these states to construct partial control and FD signals. Next, the reduced-order CDU is designed to generate the other control and FD signals. Compared with existing results based on full-order CDU, the proposed results lead to less conservatism and reduce design complexity. The switching law is constructed in the frame of persistent dwell time (PDT) switching. A novel switching number constraint condition is introduced, which further relaxes the restrictions on the dwell time of switching processes of PDT. The less restriction on dwell time degrades the performance requirement of each subsystem and upgrades the degree of freedom for switching law design. Based on the proposed results, a class of nonweighted performance indexes is introduced to characterize the fault sensitivity and robustness. Finally, the effectiveness of the proposed method is illustrated by an example.     

具有随机丢包的一类网络控制系统的故障检测

针对存在随机数据包丢失的网络环境,研究了一类网络控制系统的故障检测问题.考虑随机丢包同时发生在传感器与控制器以及控制器与执行器之间,将网络控制系统建模为含有4个模态的马尔可夫跳变线性系统.基于此类模型,构造了系统的残差发生器,相应的故障检测问题转化为H∞滤波问题.利用马尔可夫跳变线性系统理论,设计了故障检测滤波器,使得...     

Fault detection for a class of uncertain nonlinear Markovian jump stochastic systems with mode-dependent time delays and sensor saturation

This paper considers the problem of robust H_∞ fault detection for a class of uncertain nonlinear Markovian jump stochastic systems with mode-dependent time delays and sensor saturation. We aim to design a linear mode-dependent H_∞ fault detection filter that ensures, the fault detection system is not only stochastically asymptotically stable in the large, but also satisfies a prescribed H_∞-norm level for all admissible uncertainties. By using the Lyapunov stability theory and generalised Itô formula, some novel delay-dependent sufficient conditions in terms of linear matrix inequality are proposed to guarantee the existence of the desired fault detection filter. Explicit expression of the desired mode-dependent linear filter parameters is characterised by matrix decomposition, congruence transformation and convex optimisation technique. Sector condition method is utilised to deal with sensor saturation, a definite relation of sector condition parameters with fault detection system robustness against disturbances and sensitivity to faults is put forward, and weighting fault signal approach is employed to improve the performance of the fault detection system. A simulation example and an industrial nonisothermal continuous stirred tank reactor system are utilised to verify the effectiveness and usefulness of the proposed method.     

Fault detection for networked control systems subject to quantisation and packet dropout

This article addresses the stochastic fault detection (SFD) problem in finite-frequency domain for a class of networked control systems (NCSs) with respect to signal quantisation and data packet dropout. Considering a logarithmic quantiser and Markovian packet dropout, the NCS is modelled as a Markov jump linear system (MJLS) with quantisation error. Further, a new definition of finite-frequency stochastic H _? index is given, which gives a measurement of sensitivity. Subsequently, sufficient conditions are derived to guarantee that the MJLS can achieve such a performance. By virtue of the obtained conditions, the fault detection filters (FDFs) are designed in finite-frequency domain, which are valid in characterising the disturbance attenuation performance and finite-frequency fault sensitivity performance. Finally, a simulation example is given to illustrate the method and its effectiveness.     

Passivity analysis of stochastic delayed neural networks with Markovian switching

In this paper, the problem of passivity analysis is investigated for a class of stochastic delayed neural networks with Markovian switching. By applying Lyapunov functional and free-weighting matrix, delay-dependent/independent passivity criteria are presented in terms of linear matrix inequalities. The results herein include existing ones for neural networks without Markovian switching as special cases. An example is given to demonstrate the effectiveness of the proposed criteria.     

Integrated fault detection and control for two‐dimensional Markovian jump systems

This paper is concerned with the problem of integrated fault detection and control for a class of two‐dimensional (2D) discrete‐time Markovian jump systems. The mathematical model of 2D Markovian jump systems is established upon the well‐known Roesser model, and a faults detection filter/controller is proposed to detect faults and meet some control specifications simultaneously. In this strategy, it takes into account both the fault detection objective and the control objective simultaneously through certain performance levels. The integrated design problem is then formulated as a multi‐objective optimization problem, which is nonconvex in essence. Furthermore, a two‐step algorithm is developed to solve this nonconvex problem. Sufficient conditions for existence of the desired fault detection filter/controller are established in terms of LMIs. A numerical example is used to demonstrate the effectiveness of the proposed method.     

A parameter-varying fault detection filter design approach for polytopic uncertain linear systems

This paper is concerned with the robust fault detection (FD) problem for a class of polytopic uncertain linear systems driven by a Wiener process. It is assumed that the state matrix is affinely depending on unknown but bounded time-varying parameters. A switching mechanism is introduced to construct the robust FD filter with variable gains and to improve the FD performances which are inherent in the traditional FD filter with fixed gains. Finally, the filter design problem is formulated as a feasibility problem in terms of solving linear matrix inequalities (LMIs), and an example is presented to illustrate the proposed methodology.     

Fault detection for nonlinear networked control systems with stochastic interval delay characterisation

In this article, we are concerned with the fault detection (FD) problem for nonlinear networked control systems (NCSs) with network-induced stochastic interval delay. By employing the information of probabilistic distribution of networked-induced time-varying delay, nonlinear constraint and the FD filter, nonlinear stochastic delay system model is established. Based on the obtained nonlinear model, utilising FD filter as residual generator, FD of nonlinear NCSs is formulated as nonlinear H _∞-filtering problem. Especially, the stochastic stability and prescribed H _∞ attenuation level are achieved using the Lyapunov functional approach, the desired FD filter is constructed in terms of certain linear matrix inequalities, which depends on not only nonlinear level but also on delay-interval and delay-interval-occurrence-rate. Numerical examples are provided to illustrate the effectiveness and applicability of proposed technique.     

Stability of stochastic Markovian jump neural networks with mode-dependent delays

In this paper, the problem of stability analysis for a general class of uncertain stochastic neural networks with Markovian jumping parameters and mixed mode-dependent delays is considered. By the use of a new Markovian switching Lyapunov–Krasovskii functional, delay-dependent conditions on mean square asymptotic stability are derived in terms of linear matrix inequalities. Numerical examples are given to illustrate the effectiveness of the proposed approach.     

Simultaneous fault estimation for uncertain Markovian jump systems subjected to actuator degradation

This paper investigates simultaneous estimations of states, system fault, and sensor fault for a class of Markovian jump systems, resorting to the design of a robust observer, in which both the external disturbance and actuator degradation are taken into consideration. The loss‐of‐effectiveness, stuck, and outage fault cases are involved, and the considered Markovian jump system is assumed to possess generally uncertain transition rates, which can generalize the traditional bounded uncertain transition rates and partially unknown transition rates. In particular, an augmented system whose states cover the original states, system fault, and sensor fault is constructed. Then, a novel adaptive observer is presented with time‐varying gain matrices and parameter accommodation being injected to deal with the disturbance and actuator degradation. Finally, a practical example is shown to demonstrate the high efficiency of the proposed method as well as its advantages over some existing results.     

Fault detection for a class of networked nonlinear systems subject to imperfect measurements

This paper addresses the problem of fault detection (FD) for discrete-time networked systems with global Lipschitz conditions and imperfect measurements. By using Bernoulli stochastic variables and a switching signal, a unified model is proposed to describe four kinds of imperfect measurements, that is, access constraints, time delays, packet dropouts, and signal quantization. We aim to design a fault detection filter (FDF) such that, for all external disturbances and imperfect measurements, the error between the residual and fault is made as small as possible. The addressed FD problem is then converted into an auxiliary H _∞ filtering problem for discrete-time stochastic switched systems with multiple time-varying delays. By applying Lyapunov-Krasovskii approach, a sufficient condition for the existence of the FDF is derived in terms of certain linear matrix inequalities (LMIs). When these LMIs are feasible, the explicit expression of the desired FDF can also be characterized. A numerical example is exploited to show the effectiveness of the results obtained.     

Fault detection for singular switched linear systems with multiple time-varying delay in finite frequency domain

This paper deals with the problem of the fault detection (FD) for continuous-time singular switched linear systems with multiple time-varying delay. In this paper, the actuator fault is considered. Besides, the systems faults and unknown disturbances are assumed in known frequency domains. Some finite frequency performance indices are initially introduced to design the switched FD filters which ensure that the filtering augmented systems under switching signal with average dwell time are exponentially admissible and guarantee the fault input sensitivity and disturbance robustness. By developing generalised Kalman–Yakubovic–Popov lemma and using Parseval's theorem and Fourier transform, finite frequency delay-dependent sufficient conditions for the existence of such a filter which can guarantee the finite-frequency H_? and H_∞ performance are derived and formulated in terms of linear matrix inequalities. Four examples are provided to illustrate the effectiveness of the proposed finite frequency method.     

奇异Markov跳跃系统的鲁棒故障检测

研究受L2有界未知输入影响的一类奇异Markov跳跃系统的鲁棒故障检测问题。采用基于观测器的故障检测滤波器（FDF）作为残差产生器,将故障检测滤波器的设计归结为随机意义下的H∞滤波问题。推导并证明了问题可解的充分条件,并通过求解线性矩阵不等式得到了故障检测滤波器参数矩阵的解。算例验证了所给算法的有效性。     

Event-triggered filtering and fault estimation for nonlinear systems with stochastic sensor saturations

This paper is concerned with the filtering problem for a class of nonlinear systems with stochastic sensor saturations and event-triggered measurement transmissions. An event-triggered transmission scheme is proposed with hope to ease the traffic burden and improve the energy efficiency. The measurements are subject to randomly occurring sensor saturations governed by Bernoulli-distributed sequences. Special effort is made to obtain an upper bound of the filtering error covariance in the presence of linearisation errors, stochastic sensor saturations as well as event-triggered transmissions. A filter is designed to minimise the obtained upper bound at each time step by solving two sets of Riccati-like matrix equations, and thus the recursive algorithm is suitable for online computation. Sufficient conditions are established under which the filtering error is exponentially bounded in mean square. The applicability of the presented method is demonstrated by dealing with the fault estimation problem. An illustrative example is exploited to show the effectiveness of the proposed algorithm.     

Stabilization of stochastic delayed neural networks with Markovian switching

In this paper, the mean square exponential stabilization problem is investigated for a class of stochastic delayed neural networks with Markovian switching. After proposing an exponential stability condition, our attention is focused on the design of a state feedback controller such that the stochastic delayed neural networks with Markovian switching is exponentially stable in mean square. Several stabilization criteria, delay‐independent and delay‐dependent ones, which are expressed in terms of a set of linear matrix inequalities (LMIs), are proposed to stabilize the stochastic delayed neural networks with Markovian switching exponentially. The usefulness and applicability of the developed results are illustrated by means of two numerical examples. Copyright © 2008 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society