Distributed fault diagnosis of networked dynamical systems with time-varying topology

In this paper, the distributed fault diagnosis (DFD) of networked dynamical systems with time-varying connected topologies, e.g., wireless sensor networks in harsh environments, is considered. Specifically, two essential problems are focused on, which are faced in extending the the decomposition-based adaptive DFD approach to such topology-varying systems. The problems introduced by the time-varying topologies are, respectively, decomposition schemes deterioration and pre-training difficulties. The causes of the two problems are detailed and addressed in our work. First, for the decomposition schemes deterioration problem, a multi-agent dynamics-based online distributed decomposition algorithm are developed, so that a decent decomposed network structure for such topology-varying network can be maintained. Second, to alleviate the pre-training difficulties in topology-varying systems, a fault detection method is proposed, which avoids the need for pre-training. The distributed decomposition algorithm is proved to converge in finite steps, and the proposed fault detection method is verified both theoretically and experimentally.     

An alternative data-driven fault detection scheme for dynamic processes with deterministic disturbances

This paper proposes an alternative fault detection (FD) scheme, in which the so-called residual signals are generated by means of a projection of process input data. This is the major difference to the existing model-based and data-driven FD schemes, where residual generator is realized based on the process input and output relationship/dynamics. Moreover, this way of residual generation avoids the parameter identification procedure and also allows us to address deterministic disturbances (unknown inputs), which be paid often less attention by data-driven FD methods. In this fashion, the FD issue reduces to detect change of a random matrix. Since it is difficult to directly measure this change, so the trace of a matrix is adopted as the evaluation function. Furthermore, the threshold can be set by considering the boundedness of disturbance. The effectiveness of the proposed method is verified by a simulation study on an inverted pendulum system.     

Optimal residual generation for fault detection in linear discrete time-varying systems with uncertain observations

This work deals with the problem of optimal residual generation for fault detection (FD) in linear discrete time-varying (LDTV) systems subject to uncertain observations. By introducing a generalized fault detection filter (FDF) with four parameter matrices as the residual generator, a novel FDF design scheme is formulated as two bi-objective optimization problems such that the sensitivity of residual to fault is enhanced and the robustness of residual to unknown input is simultaneously strengthened. A generalized operator based optimization approach is proposed to deduce solutions to the corresponding optimization problems in operator forms, where the related H_∞/H_∞ or $H_{?}/H_{\infty }$ FD performance index is maximized. With the aid of the addressed methods, the connections among the derived solutions are explicitly announced. The parameter matrices of the FDF are analytically derived via solving simple matrix equations recursively. It is revealed that our proposed results establish an operator-based framework of optimal residual generation for some kinds of linear discrete-time systems. Illustrative examples are given to show the applicability and effectiveness of the proposed methods.     

Event-triggered fault estimation and fault-tolerant control for networked control systems

This paper is concerned with the event-triggered fault estimation and fault-tolerant control for continuous-time dynamic systems subject to system fault and external disturbance under network environment. Firstly, based on the event-triggered sampling, a fault diagnosis observer is constructed to estimate both the system state and the system fault simultaneously, and a multi-objective constraint is established to guarantee the estimation accuracy. Based on the estimated system state and fault signal, a fault-tolerant controller is proposed to compensate the influence of occurred faults and maintain the system performance. The event-triggered scheme and the fault-tolerant controller are co-designed to guarantee the required performance of faulty system and reduce the consumption of communication resources. Finally, simulation results of an F-404 aircraft engine system are provided to demonstrate the effectiveness of the proposed method.     

Observer-based fault detection for uncertain nonlinear systems

This paper addresses L₂ observer-based fault detection issues for a class of nonlinear systems in the presence of parametric and dynamic uncertainties, respectively. To this end, three different types of uncertain affine nonlinear system models studied in this paper are described first. Then, the integrated design schemes of L₂ observer-based fault detection systems are derived with the aid of Hamilton–Jacobi inequalities (HJIs), respectively. Numerical examples are also provided in the end to demonstrate the effectiveness of the proposed results.     

Active fault tolerant control scheme for aircraft with dissimilar redundant actuation system subject to hydraulic failure

In this paper, an active fault tolerant control (AFTC) scheme is proposed for more electric aircraft (MEA) equipped with dissimilar redundant actuation system (DRAS). The effect of various fault/failure of hydraulic actuator (HA) on the system performance is analyzed in this work. In nominal condition, the state feedback control law is designed for primary control surfaces. In the presence of fault/failure of certain HA, control allocation (CA) scheme together with integral sliding mode controller (ISMC) is retrofitted with existing control law and engaged the secondary (redundant) actuators into the loop. A modified recursive least square (RLS) algorithm is proposed to identify the parametric faults in HA and to measure the effectiveness level of the actuator. In an event of failure of all HA’s in the system, electro hydraulic actuators (EHA) are taken in loop to bring the system back to its nominal operation. In order to stabilize the closed-loop dynamics of HA and EHA, fractional order controllers are designed separately for each actuator. Simulations on the lateral directional model of aircraft demonstrated the effectiveness of the proposed scheme as compared to the existing methods in the literature.     

A LPV modeling of turbocharged spark-ignition automotive engine oriented to fault detection and isolation purposes

This paper illustrates the derivation of a linear parameter varying (LPV) model approximation of a turbocharged Spark-Ignition (SI) automotive engine and its usage in designing a model-based fault detection and isolation (FDI) scheme. The LPV approximation is derived from a detailed nonlinear mathematical model of the engine on the basis of the well known Jacobian approach. The resulting LPV representation is then exploited for synthesizing a bank of LPV-FDI $H_{\infty }/H_{?}$ Luenberger observers. Each observer is in charge of detecting a particular class of fault and is designed for having low sensitivity to all other exogenous inputs so as to allow an effective fault isolation. The adopted FDI scheme is gain-scheduled and exploits a set of engine variables, assumed to be measurable on-line, as a scheduling parameters. The goodness of the LPV approximation of the engine model and the effectiveness of the LPV-FDI architecture are demonstrated by several numerical simulations.     

Event-triggered fault estimation and sliding mode fault-tolerant control for a class of nonlinear networked control systems

This paper is concerned with integrated event-triggered fault estimation (FE) and sliding mode fault-tolerant control (FTC) for a class of discrete-time Lipschtiz nonlinear networked control systems (NCSs) subject to actuator fault and disturbance. First, an event-triggered fault/state observer is designed to estimate the system state and actuator fault simultaneously. And then, a discrete-time sliding surface is constructed in state-estimation space. By the use of a reformulated Lipschitz property and delay system analysis method, the sliding mode dynamics and state/fault error dynamics are converted into a unified linear parameter varying (LPV) networked system model by taking into account the event-triggered scheme, actuator fault, external disturbance and network-induced delay. Based on this model and with the aid of Lyapunov–Krasovskii functional method, a delay-dependent sufficient condition is derived to guarantee the stability of the resulting closed-loop system with prescribed H_∞ performance. Furthermore, an observed-based sliding mode FTC law is synthesized to make sure the reachability of the sliding surface. Finally, simulation results are conducted to verify the effectiveness of the proposed method.     

Evaluation of fault diagnosability for networked control systems subject to missing measurements

The evaluation of fault diagnosability for networked control systems subject to missing measurements is addressed in this paper. In particular, the missing probability is assumed to be affected by norm-bounded uncertainties. By considering noise and missing measurements, the quantitative diagnosability problem is investigated via the principle of weight. To quantify the effect of uncertain missing probabilities, an interval criterion is presented and uncertainty ratio is defined. Furthermore, a novel method is proposed to alleviate the computation task of evaluating fault diagnosability. The effectiveness of the proposed method is verified by a numerical example.     

Design and stability analysis of multivariate extremum seeking with Newton method

This paper proposes a multivariate extremum seeking with the Newton method (ES-NM) to improve the control performance for multivariable static and dynamic systems. The structure of the proposed ES-NM is designed to speed up the convergence of the scheme without increasing the oscillation. The influence of unknown Hessian matrix on the convergence speed existed in conventional methods is effectively eliminated in the proposed ES-NM approach. The stability analysis of the proposed ES-NM is given in detail for static and dynamic systems. Comparisons to the existing Gradient based extremum seeking control (ESC) and the Newton based ESC reveal that the proposed ES-NM has a higher probability of improving the convergence speed as well as reducing the chattering performance. Simulation results show advantages of the proposed ES-NM by comparing the multivariate Gradient based and Newton based ESC.     

Distributed Kalman consensus filter with event-triggered communication: Formulation and stability analysis

This paper deals with the distributed estimation problem for networked sensing system with event-triggered communication schedules on both sensor-to-estimator channel and estimator-to-estimator channel. Firstly, an optimal event-triggered Kalman consensus filter (KCF) is derived by minimizing the mean squared error of each estimator based on the send-on-delta triggered protocol. Then, the suboptimal event-triggered KCF is proposed in order to reduce the computational complexity in covariance propagation. Moreover, the formal stability analysis of the estimation error is provided by using the Lyapunov-based approach. Finally, simulation results are presented to demonstrate the effectiveness of the proposed filter.     

Subset selection for visualization of relevant image fractions for deep learning based semantic image segmentation

Semantic image segmentation is a challenging problem from image processing where deep convolutional neural networks (CNN) have been applied with great success in the recent years. It deals with pixel-wise classification of an input image, dividing it into regions of multiple object classes. However, CNNs are opaque models. Given a trained CNN, it is hard to tell which information encoded in the input image is important for the network to perform segmentation. Such information could be useful to judge whether a trained network learned to segment in a plausible way or how its performance can be improved.For a trained CNN, we formulate an optimization problem to extract relevant image fractions for semantic segmentation. We try to identify a subset of pixels that contain the relevant information for the segmentation of one selected object class. In experiments on the Cityscapes dataset, we show that this is an easy way to gain valuable insight into a CNN trained for semantic segmentation. Looking at the relevant image fractions, we can identify possible limits of a trained network and draw conclusions about possible improvements.     

Robust simultaneous fault detection and control for a class of nonlinear stochastic switched delay systems under asynchronous switching

This paper concerns the simultaneous fault detection and control (SFDC) problem for a class of nonlinear stochastic switched systems with time-varying state delay and parameter uncertainties. The switching signal of detector/controller unit (DCU) is assumed to be with switching delay, which results in the asynchronous switching between the subsystems and DCU. By constructing a switching strategy depending on the state and switching delays, new sufficient conditions expressed by a set of linear matrix inequalities (LMIs) is derived to design DCU gains. This problem is formulated as an H_∞ optimization problem and both mean square exponential stability and fault detection of augmented system are considered. A numerical example is finally exploited to verify the effectiveness and potential of the achieved scheme.     

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

This paper investigates the problem of event-triggered fault detection filter design for nonlinear networked control systems with both sensor faults and process faults. First, Takagi–Sugeno (T–S) fuzzy model is utilized to represent the nonlinear systems with faults and disturbances. Second, a discrete event-triggered communication scheme is proposed to reduce the utilization of limited network bandwidth between filter and original system. At the same time, considering network-induced delays and event-triggered scheme, a novel T–S fuzzy fault detection filter is constructed to generate a residual signal, which has nonsynchronous premise variables with the original T–S fuzzy system. Then, the fuzzy Lyapunov functional based approach and the reciprocally convex approach are developed such that the obtained sufficient conditions ensure that the fuzzy fault detection system is asymptotically stable with H_∞ performance and is less conservative. All the conditions are given in terms of linear matrix inequalities (LMIs), which can be solved by LMI tools in MATLAB environment. Finally, a numerical example is provided to demonstrate the effectiveness of the proposed results.     

Stability analysis of linear continuous-time delay-difference systems with multiple time-delays

This paper is concerned with the stability analysis of linear continuous-time delay-difference systems with multiple delays. Firstly, a new method for testing the L₂-exponential stability of the considered system is proposed, which is easier to use than the one in the existing literature. In view of the conservatism and the complexity of the obtained stability conditions in the existing literature, a complete Lyapunov–Krasovskii functional (LKF) is constructed by analyzing the relationship among the multiple delays. Sufficient conditions for both L₂-exponential stability and exponential stability are then derived based on the constructed LKFs, which are delay-independent. Exponential convergence rate for the considered system is also investigated by a new method, which is shown to be equivalent to the existing approach by using weighted LKFs. Robust stability under parameter uncertainties is also investigated. Numerical examples are provided to demonstrate the effectiveness and less conservativeness of the proposed method.