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 detection and isolation of multi-variate time series data using spectral weighted graph auto-encoders

Fault or anomaly detection is one of the key problems faced by the chemical process industry for achieving safe and reliable operation. In this study, a novel methodology, spectral weighted graph autoencoder (SWGAE) is proposed, wherein, the problem of anomaly detection is addressed with the help of graphs. The proposed approach entails the following key steps. Firstly, constructing a spectral weighted graph, where each time step of a process variable in the multivariate time series dataset is modelled as a node in an appropriately tuned moving window. Subsequently, we propose to monitor the weights of the edges between two nodes that make a connection. The faulty instances are identified based on the discrepancy in the weight pattern compared to normal operating data. To this end, once the weights are determined, they are fed to the auto-encoder network, where reconstruction loss is calculated, and faults are identified if the reconstruction loss exceeds a threshold. Further, to make the proposed approach comprehensive, a fault isolation methodology is also proposed to identify the faulty nodes once the faulty variables are identified. The proposed approach is validated using Tennessee-Eastman benchmark data and pressurized heavy water nuclear reactor real-time plant data. The results indicate that the SWGAE method, when compared to the other state-of-the-art methods, yielded more accurate results in correctly detecting faulty nodes and isolating them.     

Robust actuator fault diagnosis scheme for satellite attitude control systems

In this paper, a robust actuator fault diagnosis scheme is investigated for satellite attitude control systems subject to model uncertainties, space disturbance torques and gyro drifts. A nonlinear unknown input observer is designed to detect the occurrence of any actuator fault. Subsequently, a bank of adaptive unknown input observers activated by the detection results are designed to isolate which actuator is faulty and then estimate of the fault parameter. Fault isolation is achieved based on the well known generalized observer strategy. The simulation on a closed-loop satellite control system with time-varying or constant actuator faults in the form of additive and multiplicative unknown dynamics demonstrates the effectiveness of the proposed robust fault diagnosis strategy.     

Detecting faulty edges of complex dynamical networks based on compressive sensing

As an important technology to improve network reliability, fault diagnosis has gained wide attention in complex dynamical networks. However, few studies focused on detecting the structure of broken edges when faults occur. In this paper, due to the natural sparsity of complex dynamical networks, a completely data-driven method based on compressive sensing is established to detect the structure of faulty edges from limited measurements. The least absolute shrinkage and selection operator algorithm is applied to solve the reconstruction problem. In addition, the method is also applicable to multilayer networks. The faulty edges in both the intralayer network and the interlayer network can be fully identified. Compared with other methods, the main advantages of the proposed method lie in two aspects. First, the structure of faulty edges can be obtained directly with limited measurements. Second, the proposed method is less time consuming and more efficient due to less data processing. Numerical simulations involving single-layer, multilayer and real-world complex dynamical networks are given to demonstrate the accuracy of detecting the structure of faulty edges from the proposed method.     

Fault-tolerant control for a class of linear interconnected hyperbolic systems by boundary feedback

This paper considers a fault-tolerant control problem for a class of interconnected linear hyperbolic partial differential equation systems. Both subsystem faults and coupling faults are considered. Firstly, the well-posedness of the faulty system is analyzed by using semigroup theory. Secondly, for the fault-free case, a stabilizing boundary feedback control based on small-gain theorem is proposed. Consequently, in the presence of faults, fault recoverability conditions are established that maintain the stability of the faulty systems. The fault-tolerant control strategies are also provided. A heat exchanger example is taken to illustrate the effectiveness and practicality of the proposed methods.     

Stable image representation based stability performance monitoring and recovery for feedback control systems

This paper proposes a scheme to achieve real-time stability performance monitoring (SPM) and designs performance recovery controllers (PRCs) for feedback control systems with multiplicative faults. To be specific, a stability performance indicator is presented with the aid of stable image representation (SIR) for multiplicative faults through coprime factorization techniques. On this basis, a systematically hierarchical SPM and PRC scheme is proposed with three thresholds derived to evaluate the performance degradation degree. Subsequently, an integrated model-based and data-driven SIR-based PRC is presented to recover system stability performance. The embedded PRC parameters are adaptive to system variations by means of identifying the SIR of the faulty plant through a dual parity vector. Besides, some QR-decomposition based data-driven techniques are provided for the implementation of PRC to improve computation efficiency. Finally, the effectiveness of the proposed approach is demonstrated on a boost circuit model.     

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.     

Sampled observer-based adaptive output feedback fault-tolerant control for a class of strict-feedback nonlinear systems

This paper studies the sampled outputs-based adaptive fault-tolerant control problem for a class of strict-feedback uncertain nonlinear systems, where the nonlinear functions are allowed to include the unmeasured system states. Within the framework, a sampled output observer is introduced to jointly estimate the system states and parameters. By combining the estimated states and the supervisory switching strategy, an adaptive fault-tolerant controller is designed to achieve the desirable tracking performance. By using Lyapunov stability theory, it is proved that all the signals of the closed-loop systems are bounded and the tracking error converges to an adjustable neighbourhood of the origin eventually both in the fault free and faulty cases. Especially, if the outputs are available all the time, the proposed output feedback fault-tolerant control method can ensure the tracking error satisfy the prescribed performance bounds regardless of the faults. Finally, two examples are used to illustrate the effectiveness of the proposed method.     

Robust fault tolerant tracking controller design for a VTOL aircraft

This paper deals with the fault tolerant control (FTC) design for a Vertical Takeoff and Landing (VTOL) aircraft subject to external disturbances and actuator faults. The aim is to synthesize a fault tolerant controller ensuring trajectory tracking for the nonlinear uncertain system represented by a Takagi–Sugeno (T–S) model. In order to design the FTC law, a proportional integral observer (PIO) is adopted which estimate both of the faults and the faulty system states. Based on the Lyapunov theory and ?₂ optimization, the trajectory tracking performance and the stability of the closed loop system are analyzed. Sufficient conditions are obtained in terms of linear matrix inequalities (LMI). Simulation results show that the proposed controller is robust with respect to uncertainties on the mechanical parameters that characterize the model and secures global convergence.     

A practical propagation path identification scheme for quality-related faults based on nonlinear dynamic latent variable model and partitioned Bayesian network

In this paper, a practical technology or solution of quality-related fault diagnosis is provided for nonlinear and dynamic process. Unlike traditional data-based fault diagnosis methods, the alternative approach is focused more on identifying the propagation path that combines diagnostic information and process knowledge. The new method addresses the quality-related fault detection issue with developed nonlinear dynamic latent variable model for extracting nonlinear latent variables that exhibit dynamic correlations, then the advantage of relative reconstruction based contribution approach is followed to analyze the potential root-cause variables. Meanwhile, a new partitioned Bayesian network methodology is proposed for propagation path identification of quality-related faults. Finally, the whole proposed framework is applied to a real hot strip mill process, where the effectiveness is further demonstrated from real industrial data.     

Adaptive fault-tolerant shape control for nonlinear Lipschitz stochastic distribution systems

This paper addresses the problem of adaptive fault estimation and fault-tolerant control for a class of nonlinear non-Gaussian stochastic systems subject to time-varying loss of control effectiveness faults. In this work, time-varying faults, Lipschitz nonlinear property and general stochastic characteristics are taken into consideration in a unified framework. Instead of using the system output signal, the output distribution is adopted for shape control. Both the states and faults are simultaneously estimated by an adaptive observer. Then, a fault tolerant shape controller is designed to compensate for the faults and realize stochastic output distribution tracking. Both the fault estimation and the fault tolerant control schemes are designed based on linear matrix inequality (LMI) technique. Satisfactory performance has been obtained for a numerical simulation example. Furthermore the proposed scheme is successfully tested in a case study of particle size distribution control for an emulsion polymerization reactor.     

基于神经网络的汽油机故障诊断的专家系统

汽油发动机出现故障的机率较高,一般占整车故障的４０％左右,研究汽油发动机故障诊断专家系统,可以及时准确地对发动机技术状况做出判断,指导调整其技术状态,这无疑增加了汽车使用的可靠性,经济性和安全性。人工神经网络是基于数值计算和知识处理系统。针以地传统专家系统在处理故障诊断中的不足,提出了将人工神经网络技术与专家系统融合的模型,并将此模型应用到汽油机故障诊断中。     

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.     

Fuzzy gain-scheduled active fault-tolerant control of a wind turbine

Advanced fault detection and accommodation schemes are required for ensuring efficient and reliable operation of modern wind turbines. This paper presents a novel approach in designing a fault detection and diagnosis (FDD) and fault-tolerant control (FTC) scheme for a wind turbine using fuzzy modeling, identification and control techniques. First, an improved gain-scheduled proportional-integral (PI) control system based on fuzzy gain scheduling (FGS) technique for multi-input and multi-output wind turbine system is designed. Then, to accommodate sensor faults and based on a signal correction algorithm, an active fault-tolerant control system (AFTCS) is developed as an extension of the gain-scheduled PI control system. The AFTCS exploits the fault information from a model-based FDD scheme developed using fuzzy modeling and identification method. The proposed schemes are evaluated by a series of simulations on a well-known large off-shore wind turbine benchmark in the presence of wind turbulences, measurement noises, and different realistic fault scenarios. All results indicate high effectiveness and robustness of the designed control systems in both fault-free and faulty operations of the wind turbine.     

Identification and stabilization of faulty Boolean control networks

Fault-tolerant control is a fundamental branch in the modern control theory, and has wide applications such as aerospace, automotive technology and nuclear engineering. Particularly, the study of faulty Boolean control networks (BCNs) is meaningful to the disease treatment. This paper focuses on both stuck-at fault and bridging fault in BCNs, and investigates the identification and stabilization of BCNs subject to these two faults. The basic mathematical tool is semi-tensor product (STP) of matrices, which is used to determine the algebraic formulation of faulty BCNs. Through the construction of invariant sets corresponding to the faulty nodes, the relations between these two faults and state transition matrices are presented, which is helpful to identify the faulty nodes. In addition, the robust stabilization of BCNs subject to these two faults is discussed and several new criteria are derived. Finally, the obtained results are applied to analyze the stabilization of oxidative stress response pathways.     

High-gain observer with sliding mode for nonlinear state estimation and fault reconstruction

This paper develops a high gain observer with multiple sliding modes for simultaneous state and fault estimations for MIMO nonlinear systems. The novelty lies in the observer design that employs the combination of high-gain observer and sliding mode observer. The proposed observer does not impose the small-Lipschitz-constant condition on the system nonlinearity. By imposing a structural assumption on the nonlinear fault distribution matrix, the observability of the faults/unknown inputs w.r.t. the outputs is safeguarded and sliding modes are utilized for their reconstruction. The reconstruction of the faults from the sliding mode only relies on the output estimation error and thus can be implemented online together with the state estimation. Finally, an application to flexible joint robotic arm is used to illustrate the proposed method.     

Adaptive data-driven fault-tolerant control for nonlinear systems: Koopman-based virtual actuator approach

This paper proposes an adaptive data-driven fault-tolerant control scheme using the Koopman operator for unknown dynamics subjected to nonlinearities, time-varying loss of effectiveness, and additive actuator faults. The main objective of this method is to design a virtual actuator to hide actuator faults from the view of the system’s nominal controller without having any prior knowledge about the system’s underlying dynamics. The designed virtual actuator is placed between the faulty plant and the nominal controller of the system to keep the dynamical system’s performance consistent before and after the occurrence of actuator faults. Based on the Koopman operator theory, an equivalent Koopman predictor is first obtained using the process data only, without knowing the governing equations of the underlying dynamics. Koopman operator is an infinite-dimensional, linear operator which takes the nonlinear process data into an infinite-dimensional feature space where the dynamic data correlations have linear behavior. Next, based on the approximated system’s Koopman operator, a virtual actuator is designed and implemented without knowing the system’s nominal controller. Needless to use a separate fault detection, isolation, and identification module to perform fault-tolerant control, the current method leverages the adaptive framework to keep the system’s desired performance in facing time-varying additive and loss of effectiveness actuator faults. Finally, the approach’s efficacy is demonstrated using simulation on a two-link manipulator benchmark, and a comparison study is presented.