Adaptive Memory-Based Event-Triggered Fault Detection for Networked Stochastic Systems

This study is concerned with the event-triggered fault detection (FD) filter design problem for the discrete-time non-linear systems with multi-data transmission and output quantisation. To reduce the transmission data size, the output signals are quantified by a logarithmic quantiser. Compared with the existing results, past output measurements are first introduced for packaging and transmission only when the event condition is satisfied. Based on this consideration, more data can be used by the FD filter, while using the traditional method, the FD filter can only use the current data. Finally, some simulation results show that the proposed design method can achieve a better FD performance than the existing method with current output measurements transmission.

This paper is concerned with the event-triggered fault detection filter (FDF) design for an unmanned surface vehicle (USV) under the network environment. A framework of fault detection is established for a USV subject to wave-induced disturbance and actuator failures, in which an FDF is utilized to construct a residual model and an integral-based event generator is introduced to save communication resources. Compared with the traditional instantaneous value based event-triggering scheme and periodic sampling, the proposed event-triggering mechanism can not only reduce bandwidth utilization of the network more significantly, but also get rid of the Zeno phenomenon fundamentally. The event-triggering scheme and the FDF are co-designed. Finally, the efficient performance of the proposed fault detection method based on integral-based event-triggering scheme is illustrated by simulation.

Event-triggered fault detection filter design for nonlinear networked systems via fuzzy Lyapunov functions

This paper discusses adaptive event-triggered fault detection filter design for fuzzy stochastic models with missing measurements. First, a novel event-triggered strategy is introduced, while an adaptive law is provided to adjust communication threshold dynamically. Compared with traditional event-triggered methods with fixed threshold, the proposed strategy is more effective on saving network communication resources. Second, a Bernoulli stochastic process is proposed to describe the measurement missing phenomenon, which always appears in real network environment. Then, an integrated fault detection model for fuzzy stochastic systems is constructed by taking network-induced delays, adaptive event-triggered strategy and missing measurements into account. A new method is provided to achieve mean-square asymptotical stability of residual model with one desired fault detection objective. Finally, simulation cases are introduced to verify the validity of the designed strategy.

ABSTRACTThis work focuses on the fault detection for nonlinear networked systems. A fault detection filter is designed to detect the fault signal under the effect of disturbance. The communication channel of the networked system is supposed to be influenced by the limited bandwidth and random data losses. To deal with these, event-triggered scheme and Bernoulli process are employed to describe the problems, respectively. According to the above situation, we employ a formulated random series to describe the input of fuzzy filter. Differential mean value theorem based method is presented to handle the asynchronous membership functions. Synthetically, linear matrix inequality approach is employed to obtain the stability conditions and the filter design process. At last, we verify the performance of this method acting in the simulation example.

This study investigates the problem of fault detection (FD) filter design for a class of uncertain Lipschitz systems with time delays, unknown faults, parameter uncertainties and bounded disturbances. A novel FD filter is proposed and the augmented filter system is converted to a linear parameter varying (LPV) system by the reformulated Lipschitz property. Then, two bounded real lemmas (BRLs) are given for the generated residual signal to achieve the fault sensitivity performance in low-/middle-/high-frequency domains and disturbance attenuation performance in the full-frequency domain. Next, the obtained BRLs are further converted into a set of linear matrix inequalities (LMIs) to implement the filter design. Finally, two illustrative simulations including a numerical example and Chua's circuit system clearly demonstrate the advantages of the developed FD scheme.

On design of quantized fault detection filters with randomly occurring nonlinearities and mixed time-delays

In this paper, the design problem of fault detection filter for a class of networked control system subject to Markov jump parameters, randomly occurring uncertainties, sensor faults and packet dropouts by using state feedback control is investigated. The fault detection filter is used as an optimal residual generator that generates the residual signal and also guarantees the sensitivity of residual signal to the faults. To be more specific, a suitable threshold is selected for the residual evaluation function to reduce the error between the sensor faults and the residual signal. Further, by employing a fault detection technique along with output quantization approach, a new set of sufficient conditions is derived to ensure the stochastic stabilization of the addressed system with sensor faults under a predefined $$H_{\infty }$$ performance level. Particularly, by constructing Lyapunov–Krasovskii functional and utilizing extended Wirtinger-based single integral inequality, the required sufficient conditions are derived in terms of linear matrix inequalities for getting the required result. At last, two numerical examples including high alpha research vehicle model are presented to demonstrate the usefulness of our developed results.

This paper is concerned with the problem of robust H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> fault detection for a class of Itô stochastic T-S fuzzy systems with time-varying delays and parameter uncertainties. We aim to design fuzzy-rule-independent fault detection filter, which insures the fault detection system is not only mean square asymptotically stable, but also satisfies a prescribed H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> -norm level for all admissible uncertainties. According to Lyapunov stability theory and Itô formula, some novel delay-dependent sufficient conditions in terms of linear matrix inequality (LMI) are proposed to guarantee the existence of the desired fault detection filter. Weighted fault signal approach is used to improve the performance of the fault detection system and a simulation example is employed to demonstrate the effectiveness of the proposed approach.

This paper deal with the problem of the fault detection (FD) filter design for continuous-time switched systems with actuator faults. The actuator faults and the unknown disturbances of the system are set in finite frequency domain. By using the switched Lyapunov function and the average dwell-time (ADT) techniques, efficient conditions are obtained, which can realize the residual signal sensitive to the fault and robust to the unknown disturbances. In the design of fault detection filter, we use Linear matrix inequalities (LMIs) conditions to guarantee the finite frequency H∞ and H - performance index. Finally, a practical example is provided and simulation results are conducted to demonstrate the effectiveness of the proposed approach.

This technical note deals with the problem of finite horizon H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> fault detection for linear discrete time-varying systems. Using an observer-based fault detection filter (FDF) with residual feedback as a residual generator, a new problem formulation of fault detection is first presented. Then the design of finite horizon H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> -FDF is converted into a minimum problem of certain quadratic form. Through building a relationship with Krein space projection and applying innovation analysis, a sufficient and necessary condition for the minimum is derived and a solution to the H <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">∞</sub> -FDF is obtained by recursively computing Riccai recursions. Numerical examples are given to show the effectiveness of the proposed method.

Fault detection filters design of polytopic uncertain discrete-time singular Markovian jump systems with time-varying delays

In this paper, for the purpose of improving the energy efficiency of the industrial sensor networks, we investigated the event-based H∞ filtering problem for a class of discrete-time nonlinear sensor network systems with time-varying delay, packet dropout, and multiplicative noises. Instead of traditional time-triggered communication mechanism, the event-triggered strategy is adopted in industrial sensor network, which could not only reduce the transmission frequency of the sensor measurement output, but also guarantee the prescribed filtering performance, if only the threshold in the event-triggered function is chosen suitably. The time-varying delay characteristic of systems is considered with the event-triggered strategy, which has seldom been studied due to the complexity of time-varying delay and event-triggered strategy. The most common network-induced phenomenon of packet dropout in industrial sensor network is described. The purpose is to design a filter satisfying exponentially stable and H∞ indexes. The main result is that sufficient conditions are established, guaranteeing our proposed filter satisfying filtering performance constraints, and the parameters of filter could be got through the derived linear matrix inequality (LMI), if only it is feasible. At last, the filtering approach is demonstrated by a simulation.

This paper proposes a novel full-order and reduced-order fault detection filters (FDFs) design method for singular continuous-time Markovian jump systems (SCTMJSs) with completely known transition rates (TRs). In order to strengthen the reliability of full-order and reduced-order (FDFs), we incorporates the idea of congruent transformation Lyapunov matrix to solve the convex optimization problem. The superiority of this kind of design method is that shows full-order and reduced-order FDFs can be expressed by an inequality and effective. Particularly necessary to point out that the reduced-order fault detection filter (FDF) has fewer variables and shortens the time of fault detection (FD) compared with full-order FDF. At the end of this paper, an illustrative simulation example is given to verify the effectiveness of the proposed method.

This paper introduces an input delay approach to fault detection of networked control systems (NCSs) with network-induced delay and data dropout. By using an input delay approach, the NCS is modeled as a continuous-time system with time-varying delays, where the piecewise-constant feedback control input is represented as the state with two additive delay components. A fault detection filter (FDF) is introduced to be the residual generator. Based on a new result on H ∞ performance for systems with two successive delay components in the state, a sufficient condition for the existence of the FDF is presented in terms of a nonlinear matrix inequality formulation, dependently of time delay. To make the matrix linear, an iterative procedure of linear matrix inequality (LMI) minimum problem is given. A numerical example is employed to demonstrate the validity of the proposed approach.