Parallel supervised additive and multiplicative faults detection for nonlinear process

In this paper, a novel supervised nonlinear process monitoring method named comprehensive kernel principal component regression (C-KPCR) is proposed to monitor the quality-related/unrelated additive/multiplicative faults. Firstly, mutual information is used to classify the process variables into quality-related part and quality-unrelated part. Secondly, the original variables matrix and the variables variance matrix are constructed and the data is mapped into high-dimensional feature space to deal with the nonlinear problem. Then the quality-related additive and multiplicative faults can be detected based on the regression model using original variables matrix and variables variance matrix, respectively. Afterwards, the monitoring result of quality-unrelated fault is obtained through combining the quality-unrelated information in the regression model and the quality-unrelated process variables. Finally, the effectiveness of the proposed method is demonstrated by a numerical example and the Tennessee Eastman process.     

Fault diagnosis and fault tolerant control for the non-Gaussian time-delayed stochastic distribution control system

The purpose of fault diagnosis of stochastic distribution control (SDC) systems is to use the measured input and the system output probability density functions (PDFs) to obtain the fault information of the SDC system. When the target PDF is known, the purpose of fault tolerant control of stochastic distribution control system is to make the output PDF still track the given distribution using the fault tolerant controller. However, in practice, time delay may exist in the data (or image) processing, the modeling and transmission phases. When time delay is not considered, the effectiveness of the fault detection, diagnosis and fault tolerant control of stochastic distribution systems will be reduced. In this paper, the rational square-root B-spline is used to approach the output probability density function. In order to diagnose the fault in the dynamic part of such systems, it is then followed by the novel design of a nonlinear neural network observer-based fault diagnosis algorithm. The time delay term will be deleted in the stability proof of the observation error dynamic system. Based on the fault diagnosis information, a new fault tolerant controller based on PI tracking control is designed to make the post-fault probability density function still track the given distribution, which is dependent of the time delay term. Finally, simulations for the particle distribution control problem are given to show the effectiveness of the proposed approach.     

A process monitoring and fault isolation framework based on variational autoencoders and branch and bound method

Nonlinear characteristic widely exists in industrial processes. Many approaches based on kernel methods and machine learning have been developed for nonlinear process monitoring. However, the fault isolation for nonlinear processes has rarely been studied in previous works. In this paper, a process monitoring and fault isolation framework is proposed for nonlinear processes using variational autoencoder (VAE) model. First, based on the probability graph model of VAE, a uniform monitoring index can be calculated by the probability density of observation variables. Then, the fault variables are estimated with normal variables by a missing value estimation method. The optimal fault variable set can be searched by branch and bound (BAB) algorithm. The proposed method can resolve the ”smearing effects” problem existing in traditional fault isolation methods. Finally, a numerical case and a hot strip mill process case are used to verified the proposed method.     

Adaptive fault compensation control for nonlinear uncertain fractional-order systems: Static and dynamic event generator approaches

The tracking problem of the fractional-order nonlinear systems is assessed by extending new event-triggered control designs. The considered dynamics are accompanied by the uncertain strict-feedback form, unknown actuator faults and unknown disturbances. By using the neural networks and the fault compensation method, two adaptive fault compensation event-triggered schemes are designed. Unlike the available control designs, two static and dynamic event-triggered strategies are proposed for the nonlinear fractional-order systems, in a sense that the minimum/average time-interval between two successive events can be prolonged in the dynamic event-triggered approach. Besides, it is proven that the Zeno phenomenon is strictly avoided. Finally, the simulation results prove the effectiveness of the presented control methods.     

Fault detection of discrete-time T–S fuzzy affine systems based on piecewise Lyapunov functions

This paper investigates the problem of robust fault detection for a class of discrete-time nonlinear systems, which are represented by Takagi–Sugeno (T–S) fuzzy affine dynamic models with norm-bounded uncertainties. The objective is to design an admissible fault detection filter guaranteeing the asymptotic stability of the resulting residual system with prescribed performances. It is assumed that the plant premise variables, which are often the state variables or their functions, are not measurable so that the fault detection filter implementation with state-space partition may not be synchronized with the state trajectories of the plant. Based on a piecewise quadratic Lyapunov function combined with S-procedure and some matrix inequality convexification techniques, the results are formulated in the form of linear matrix inequalities. Finally, a simulation example is provided to illustrate the effectiveness of the proposed approach.     

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.     

Nonlinear process monitoring based on generic reconstruction-based auto-associative neural network

A significant concern with statistical fault diagnosis is the large number of false alarms caused by the smearing effect. Although the reconstruction-based approach effectively solves this problem, most of them only focus on linear rather than nonlinear systems. In the present work, a generic reconstruction-based auto-associative neural network (GRBAANN) is proposed that uses the reconstruction-based approach to isolate simple and complex faults for nonlinear systems. Nevertheless, in GRBAANN, it is challenging to acquire a trivial solution for the reconstruction-based index, which is equivalent to a complex vector fixed-point problem. In this regard, the Steffensen method is employed to deal with this problem with an accelerated iterative process, which is appropriate for both single and multiple variable faults. The variable selection procedure is time-consuming but imperative for reconstruction-based approaches, with no exception to the proposed method. In order to ensure the real-time diagnosis for large-scale systems, the Sequential floating forward selection method with memory is proposed to minimize the computation time of the variable selection procedure. The effectiveness of the proposed GRBAANN scheme is illustrated through a validation example and an industrial example. Comparisons with the state-of-art methods are also presented.     

Global output feedback tracking control for switched nonlinear systems with deferred prescribed performance

In this paper, the global output feedback tracking control is investigated for a class of switched nonlinear systems with time-varying system fault and deferred prescribed performance. The shifting function is introduced to improve the traditional prescribed performance control technique, remove the constraint condition on the initial value, and make the constraint bounds have more alternative forms. To estimate the unmeasured state variables and compensate the system fault, the switched dynamic gain extended state observer is constructed, which relaxes the traditional Lipschitz conditions on the nonlinear functions. Based on the proposed observer, by constructing the new Lyapunov function and using the backstepping method, the global robust output feedback controller is designed to make the output track the reference signal successfully, and after the adjustment time, the tracking error enters into the prescribed set. The stability of the system is analyzed by the average dwell time method. Finally, simulation results are given to illustrate the effectiveness of the proposed method.     

Fault detection and diagnosis for stochastic systems via output PDFs

This paper investigates a new type of fault detection and diagnosis (FDD) problem for non-Gaussian stochastic distribution systems via the output probability density function (PDF). The PDF can be approximated by using square root B-spline expansions. In this framework, an optimal fault detection algorithm is presented by introducing the tuning parameter such that the residual is as sensitive as possible to the fault. When the fault occurs, an adaptive network parameter-updating law is designed to approximate the fault. At last, paper-making process example is given to demonstrate the efficiency of the proposed approach.     

Direct projection to latent variable space for fault detection

Partial least squares (PLSs) often require many latent variables (LVs) T to describe the variations in process variables X correlated with quality variables Y, which are obtained via the traditional nonlinear iterative PLS (NIPALS) optimal solution based on (X, Y). Total projection to latent structures (T-PLSs) performs further decomposition to extract LVs T_y directly related to Y from T, which are obtained by the PCA optimal solution based on the predicted value of Y. Inspired by T-PLS, combined with practical process characteristics, two fault detection approaches are proposed in this paper to solve problems encountered by T-PLS. Without the NIPALS, (X, Y) are projected into the latent variable space determined by main variations of Y directly. Furthermore, the structure and characteristics of several modified methods in statistical analysis are studied based on calculation procedures of solving PCA, PLS and T-PLS optimization problems, and the geometric significance of the T-PLS model is demonstrated in detail. Simulation analysis and case studies both indicate the effectiveness of the proposed approaches.     

Parameter fault detection and estimation of a class of nonlinear systems using observers

The focus of this paper is on the detection and estimation of parameter faults in nonlinear systems with nonlinear fault distribution functions. The novelty of this contribution is that it handles the nonlinear fault distribution function; since such a fault distribution function depends not only on the inputs and outputs of the system but also on unmeasured states, it causes additional complexity in fault estimation. The proposed detection and estimation tool is based on the adaptive observer technique. Under the Lipschitz condition, a fault detection observer and adaptive diagnosis observer are proposed. Then, relaxation of the Lipschitz requirement is proposed and the necessary modification to the diagnostic tool is presented. Finally, the example of a one-wheel model with lumped friction is presented to illustrate the applicability of the proposed diagnosis method.     

Fault diagnosis for a class of nonlinear uncertain systems using deterministic learning approach

In this paper, we propose a fault diagnosis (FD) approach for a class of nonlinear uncertain systems based on the deterministic learning approach (DLA). Specifically, an adaptive learning observer is constructed, in which the adaptive neural networks (NNs) are constructed to approximate the unknown system dynamics under normal and fault modes. Based on the strictly positive real (SPR) condition, the convergence of the state estimation can be guaranteed. When the system is undergoing a periodic or periodic-like (recurrent) motion, the states of the observer will also become recurrent. Thus through DLA, the partial persistent excitation (PE) condition of the associated subvectors of NNs is satisfied. By utilizing the partial (PE) condition, the uniformly completely observable (UCO) property of the identification system is analyzed and the exponential convergence condition of the identification system is derived. Under this condition, the unknown dynamics under normal and fault modes can be accurately identified along the system trajectory. And by utilizing the knowledge obtained in the identification phase, the fault can be detected in the diagnosis phase. The main attraction of this paper lies in the analytical result, which shows that the exponential convergence condition of the learning observer not only depends on the observer gain matrix, but also depends on the PE level of the regressor subvector of NN. Simulation results are included to illustrate the effectiveness of the proposed scheme.     

Fault detection strategy combining NARMAX model and Bhattacharyya distance for process monitoring

The complexity of modern chemical and petrochemical plants is becoming increasingly problematic in the recent years. At the same time, the demands to ensure safety and reliability of process operations rise. Early detection of abnormal event in complex real systems decrease maintenance cost and lead to guarantee the safety of human operators and environment. In the present work, a fault detection (FD) method which exploits the advantages of black-box modeling and statistical measure for fault detection in real chemical process as a distillation column is proposed. This technique is developed by applying the Nonlinear Auto-Regressive Moving Average with eXogenous input (NARMAX) model and Bhattacharyya distance (BD). In order to determine the NARMAX model, a real data set recorded during normal operations is used. Then, the BD is used to quantify on-line the dissimilarity between the current and reference probability distributions of the residual obtained from the NARMAX model for fault detection purposes. The ability of the proposed FD approach is demonstrated using real fault of separation unit. The obtained results indicate that the developed technique produces favorable performance compared to the conventional Cumulative Sum (CUSUM) test.     

Adaptive neural network dynamic surface control of uncertain strict-feedback nonlinear systems with unknown control direction and unknown actuator fault

In this paper, the tracking control problem of a class of uncertain strict-feedback nonlinear systems with unknown control direction and unknown actuator fault is studied. By using the neural network control approach and dynamic surface control technique, an adaptive neural network dynamic surface control law is designed. Based on the neural network approximator, the uncertain nonlinear dynamics are approximated. Using the dynamic surface control technique, the complexity explosion problems in the design of virtual control laws and adaptive updating laws can be overcome. Moreover, to solve the unknown control direction and unknown actuator fault problems, a type of Nussbaum gain function is incorporated into the recursive design of dynamic surface control. Based on the designed adaptive control law, it can be confirmed that all of the signals in the closed-loop system are semi-global bounded, and the convergence of the tracking error to the specified small neighborhood of the origin could be ensured by adjusting the designing parameters. Finally, two examples are provided to demonstrate the effectiveness of the proposed adaptive control law.     

铝电解过程故障诊断综述

铝电解是一个非线性、多耦合、时变和大时滞的工业过程体系,故障发生频率较高。本文介绍了当今铝电解故障诊断所采用的方法和国内外研究现状,以及发展趋势,为铝电解故障诊断的研究提供了理论研究基础。     

A quantitative causal diagram based optimal sensor allocation strategy considering the propagation of fault risk

Optimal sensor allocation can substantially reduce the life cycle maintenance costs of engineering systems. Considerable effort has been exerted to model the causal relationship between sensors and faults, but without considering the propagation of fault risk. In this paper, a grey relational analysis (GRA) based quantitative causal diagram (QCD) sensor allocation strategy is proposed that can take account of the influence of the propagation of fault risk. QCD is used to describe both the fault-sensor causal relationship and the fault-to-fault causal relationship. A data-driven-based GRA is applied in QCD to calculate the coefficients of the propagation of fault risk. To achieve an accurate relationship between faults and sensors, an improved quantitative analytic hierarchy process is proposed to calculate the coefficients between faults and sensors that is defined as sensor detectability in this paper. An optimal sensor allocation strategy is then developed using an improved particle swarm optimization (IPSO) algorithm under the constraint on sensor detectability to minimize fault unobservability and total cost. The proposed strategy is demonstrated by a case study on a single-phase inverter system. Compared with two other sensor allocation strategies, the results show that the proposed strategy can obtain the lowest fault unobservability of per unit cost (?0.242) for sensor allocation under the propagation of fault risk.     

A new robust fault-tolerant controller for self-repairing flight control system

A new robust fault-tolerant controller scheme integrating a main controller and a compensator for the self-repairing flight control system is discussed. The main controller is designed for high performance of the original faultless system. The compensating controller can be seen as a standalone loop added to the system to compensate the effects of fault guaranteeing the stability of the system. A design method is proposed using nonlinear dynamic inverse control as the main controller and nonlinear extended state observer-based compensator. System robustness is greatly improved by using the new configuration controller. The stability of the whole closed-loop system is analyzed. Feasibility and validity of the new controller is demonstrated with an aircraft simulation example.     

Event-triggered secure control of nonlinear multi-agent systems under sensor attacks

In this paper, a security consistent tracking control scheme with event-triggered strategy and sensor attacks is developed for a class of nonlinear multi-agent systems. For the sensor attacks on the system, a security measurement preselector and a state observer are introduced to combat the impact of the attacks and achieve secure state estimation. In addition, command filtering technology is introduced to overcome the “complexity explosion” caused by the use of the backstepping approach. Subsequently, a new dynamic event-triggered strategy is proposed, in which the triggering conditions are no longer constants but can be adjusted in real time according to the adaptive variables, so that the designed event-triggered mechanism has stronger online update ability. The measurement states are only transmitted through the network based on event-triggered conditions. The proposed adaptive backstepping algorithm not only ensures the security of the system under sensor attacks but also saves network resources and ensures the consistent tracking performance of multi-agent systems. The boundedness of all closed-loop signals is proved by Lyapunov stability analysis. Simulation examples show the effectiveness of the control scheme.