Search

In this paper, an approach to isolate the sensor and control surface/actuator failures affecting the innovation of Kalman filter was proposed and applied to an UAV dynamic model. To diagnose if the fault is a sensor fault or an actuator fault, a two-stage Kalman filter (TSKF) insensitive to actuator faults is developed. In the proposed method, sensor faults are isolated by the normalized innovation of Kalman filter. Furthermore, an adaptive linear adaptive TSKF algorithm is used to estimate the loss of control effectiveness and the magnitude of degree of stuck faults in a UAV model. Control effectiveness factors and stuck magnitudes are used to quantify faults entering control systems through actuators. In the simulations, the longitudinal and lateral dynamics of the UAV model is considered, and detection and isolation of sensor and control surface/actuator failures are examined.

In this paper we present a method for actuator fault detection and identification in Lure systems. The Lure plant is controlled by an observer-based feedback tracking controller, designed for the nominal (fault-free) system. A residual signal is constructed from measurable estimation errors associated with the nominal observer. Faults are diagnosed by on-line contrasting the residual signal trajectories against a set of values that the residuals can be shown to attain under healthy or faulty operation. These values are obtained via set-invariance analysis of the system closed-loop trajectories.

In this paper, a sliding mode based fault-tolerant control (SM-FTC) is designed, implemented and tested in a quadrotor unmanned small helicopter under the propeller damage and actuator fault conditions. Based on the concept of sliding mode control, both passive and active fault-tolerant control laws have been designed and experimentally testbed on a quadrotor UAV (unmanned aerial vehicle) testbed available at Concordia University. These two types of controllers are carried out and compared through both theoretical analyses and experimental tests on the quadrotor UAV system.

International audience; In this paper, a two-level reconfiguration action is used to design an FTC system. The objective is to adapt the corrective action accordingly to the fault severity. The first level of the procedure is triggered when the impact of the fault on the system performance is limited. At the second level a dynamic reference modification based on reference-offset governor is proposed. The idea consists of updating the controller by the new post-fault system model in first time and; in second time; modifying the set-point or reference according to the system constraints which become more strict after fault occurrence in order to avoid any actuator saturation and ensure system stability. The effectiveness of the proposed solution is illustrated by a wind turbine example subjects to actuator faults and constrained on the actuator dynamic ranges.

—This paper presents Fault-Tolerant Control (FTC) architecture for vehicle dynamics represented by an LPV (Linear Parameter Varying) model with actuator failures. The type of faults considered in this work is total or partial loss of actuator effectiveness. This method does not need any on-line fault estimation, detection or identification mechanism. The proposed approach is applied to the actuator fault tolerant control of vehicle dynamics. It is based on the resolution of LMIs , using the concept of D C -stability and involving PDLM (Parameter-Dependent Lyapunov Matrix). In order to highlight the performance of the proposed control algorithm, numerical simulations are performed.

The contribution addressed by this paper refers to the development of a new dynamic co-active neuro-fuzzy system and its application to fault detection and isolation of an evaporation station. The training of the neuro-fuzzy system is done by a hybrid learning. This is based on a fuzzy clustering algorithm to determine the number of fuzzy rules and the values of the premise parameters, and steepestdescent algorithms to basically determine the consequent parameters. The developed dynamic co-active neuro-fuzzy system is then tested in the framework of an experimental case study. This refers to the sensor and actuator fault diagnosis of an evaporation station from a sugar factory. For this purpose, an extended neuro-fuzzy generalised observer scheme is designed to generate the residuals (symptoms) in the form of the one-step-ahead prediction errors. These are then processed by a neural classifier in order to take the appropriate decision regarding the actual behaviour (normal or faulty) of the process.

Actuator fault diagnosis problem is studied for a class of nonlinear systems with relative degrees from the inputs to the outputs higher than one. This type of nonlinear systems include many mechanical systems, and their actuator fault diagnosis problem can be very challenging because of nonlinearities and high relative degrees. In this paper, in order to solve the actuator fault diagnosis problem for the considered nonlinear systems, 2nd-order and 3rd-order sliding mode differentiators are used and actuator fault diagnosis schemes are proposed to achieve fault detection and isolation. Computer simulations are carried out to compare the efficacy of the proposed fault diagnosis schemes on a laboratory 3D Crane model with noisy measurements. The fault diagnosis schemes are further tested through experiments on a laboratory 3D Crane in terms of actuator fault detection and isolation.

The objective of a robust fault reconstruction scheme is to generate an accurate reconstruction of the fault that is unaffected by disturbances. A typical method for robust fault reconstruction is to reconstruct the faults and disturbances, which is conservative and requires stringent conditions. This paper investigates and presents conditions that guarantee a fault reconstruction that rejects the effects of disturbances, which are less stringent than those of previous work. A VTOL aircraft model is used to validate the work of this paper.

The contribution addressed by this paper refers to the development of a new dynamic neuro-fuzzy system and its application to fault detection and isolation of an evaporation station. Hybrid learning based on the fuzzy c-means clustering and steepest-descent method algorithms are used to train the neuro-fuzzy system. The experimental case study refers to the sensor and actuator fault diagnosis of an evaporation station from a sugar factory. An extended neuro-fuzzy generalised observer scheme is used to generate the residuals (symptoms) in the form of the one-step-ahead prediction errors. These are then analysed by a neural classifier in order to take the appropriate decision regarding the actual behaviour of the process.

Actuator fault diagnosis is often studied under strong assumptions on available sensors: typically it is assumed that either the sensors are fault-free or there is a sufficient redundancy among the sensors. The purpose of this paper is to propose a new method for actuator fault diagnosis which is robust to sensor distortion. It does not require sensor redundancy to compensate sensor distortion. The essential assumption is that sensor distortions are strictly monotonous. Despite the nonlinear and unknown nature of distortions, such sensors still provide useful information for fault diagnosis. A numerical example is provided to illustrate the efficiency of the proposed method. Copyright © 2006 IFAC

This paper presents a computational procedure to evaluate the admissibility of actuator fault configurations (AFC) when a nonlinear predictive/optimal control with constraints is used. The procedure is based on formulating a constraint satisfaction problem from the constraints of the predictive/optimal control law including the effect of faults either through reconfiguration or accommodation. Finally, the proposed admissibility evaluation method is tested on a real application based on the predictive control of a sewer network using its nonlinear model.

Process engineering systems encountered in many risky industries (nuclear, chemical…) are complex because of their multidomain energy character. Actuators are the main elements for their control design. They need for their safety a monitoring system. The monitorability model based analysis (ability to detect and to isolate an actuator fault) is based on the structured residual analysis using analytical redundancy or covering causal path based on bond graph methodology. The present paper proposes a new approach based on flatness topology for actuator monitorability analysis. The developed approach is illustrated by a thermofluid application (a non-linear multienergy system).

This article presents the synthesis and the real tests of an actuator fault diagnosis method applied to a winding process. Based on the variation, during operating conditions, of the winding reels radius, a linear time varying model is considered to describe the dynamic behavior of this system. A stochastic detection filter is defined in order to detect and isolate actuator faults. The implementation to the process is carried out with a DSP dSpace board interfaced with Matlab/Simulink.

The emergence of low-cost microelectronics promotes the integration of different technologies to produce a local autonomous element, such as 'smart' sensors and actuators. Various requirements are needed so as to achieve smartness. One of the most important issues is the integration of self-diagnosis. For the implementation of self-diagnosis, different techniques can be employed, such as parameter estimation. The decision may depend exclusively on the type of evaluated elements. This paper addresses the implementation of a smart actuator with built-in fault diagnosis strategies. Several results are presented with respect to different scenarios in order to highlight the suggested approach.

The decentralisation of the diagnostic tasks could be achieved when the automatic and technological equipment devices will be provided with class of smart autonomous diagnostic units. In this case the functions of the central diagnostic system may be simplified. This class may be represented by smart positioner with actuator fault diagnosis. In the paper an idea of smart positioner with actuator fuzzy based fault diagnosis is presented. The scheme of the residuum vector generation based on the fuzzy models is given, as well as with the scheme of diagnostic's decision making procedure for the positioner.

This paper presents a novel robust adaptive observer based approach for the actuator fault diagnosis in dynamic systems which are expressed by a state space model and are subjected to structured model uncertainties, external disturbances and dynamic uncertainties. An adaptive observer, whose adaptive tuning part only involves the estimation of the fault, is initially constructed to demonstrate the possibility of using adaptive updating rules to estimate the fault This is followed by a description of how to select a fixed observer gain matrix in order to make the resulting adaptive observer sensitive to the fault and insensitive to the model uncertainties and external disturbances. The method is demonstrated through its application to the monitoring of a simulated distillation column.

A novel fault diagnosis method is proposed for linear stochastic systems described by autoregressive (AR) model. The method is based on likelihood decomposition. Assuming that, at some time instant τ, the fault of one of the following two types is occurs: innovation fault (actuator fault); and observation fault (sensor fault), the log-likelihood function is decomposed into two parts based on the observations before and after τ, respectively. Then, the type of the fault is determined by comparing the log-likelihoods corresponding two types of faults. Numerical examples demonstrate the usefulness of the proposed diagnosis method.

The purpose of this article is to investigate the robustness to model uncertainties of observer based fault detection and isolation. The approach is designed with a straight forward dynamic model and the observer is designed using the eigenstructure assignment technique. The faults are detected with a fixed threshold value. The example used for design and test is an electromechanical control loop which is subject to an actuator fault, a sensor fault and a load disturbance. The paper contributes with design, practical application and robustness evaluation of the method.

In the case of a non-stationary system whose the structure is described by jumping parameters, the residuals found from the difference between measured outputs and inputs model are affected by the variation of those parameters. Therefore, it becomes difficult to distinguish a sensor fault from an actuator fault or a perturbation due to parameters jumping. To bypass this difficulty, an adaptive algorithm has been developed and allows us to scan the fluctuation of model parameters. This algorithm can also give robust residuals for parameters jumping.