Wyniki wyszukiwania

1

Towards robustness in neural network based fault diagnosis

100%

Patan K., Witczak M., Korbicz J.

International Journal of Applied Mathematics and Computer Science

|

2008

|

tom 18

|

nr 4

443-454

EN

Challenging design problems arise regularly in modern fault diagnosis systems. Unfortunately, classical analytical techniques often cannot provide acceptable solutions to such difficult tasks. This explains why soft computing techniques such as neural networks become more and more popular in industrial applications of fault diagnosis. Taking into account the two crucial aspects, i.e., the nonlinear behaviour of the system being diagnosed as well as the robustness of a fault diagnosis scheme with respect to modelling uncertainty, two different neural network based schemes are described and carefully discussed. The final part of the paper presents an illustrative example regarding the modelling and fault diagnosis of a DC motor, which shows the performance of the proposed strategy.

2

Robust multisensor fault tolerant model-following MPC design for constrained systems

100%

Yetendje A., Seron M., De Doná J.

International Journal of Applied Mathematics and Computer Science

|

2012

|

tom 22

|

nr 1

211-223

EN

In this paper, a robust fault-tolerant control strategy for constrained multisensor linear systems, subject to sensor faults and in the presence of bounded state and output disturbances, is proposed. The scheme verifies that, for each sensors-estimator combination, suitable residual variables lie inside pre-computed sets and selects a more appropriate combination based on a chosen criterion. An active fault tolerant output feedback controller yields an MPC-based control law and, by means of the notion of a "tube" of trajectories, we ensure robust closed-loop exponential stability and good performance in the fault-free case and under the occurrence of abrupt sensor faults.

3

Advances in model-based fault diagnosis with evolutionary algorithms and neural networks

100%

Witczak M.

International Journal of Applied Mathematics and Computer Science

|

2006

|

tom 16

|

nr 1

85-99

EN

Challenging design problems arise regularly in modern fault diagnosis systems. Unfortunately, the classical analytical techniques often cannot provide acceptable solutions to such difficult tasks. This explains why soft computing techniques such as evolutionary algorithms and neural networks become more and more popular in industrial applications of fault diagnosis. The main objective of this paper is to present recent developments regarding the application of evolutionary algorithms and neural networks to fault diagnosis. In particular, a brief introduction to these computational intelligence paradigms is presented, and then a review of their fault detection and isolation applications is performed. Close attention is paid to techniques that integrate the classical and soft computing methods. A selected group of them is carefully described in the paper. The performance of the presented approaches is illustrated with the use of the DAMADICS fault detection benchmark that deals with a valve actuator.

4

Optimization and pole assignment in control system design

100%

Chu E. K.

International Journal of Applied Mathematics and Computer Science

|

2001

|

tom 11

|

nr 5

1035-1053

EN

Some elementary optimization techniques, together with some not so well-known robustness measures and condition numbers, will be utilized in pole assignment. In particular, ''Method 0'' by Kautsky et al. (1985) for optimal selection of vectors is shown to be convergent to a local minimum, with respect to the condition number . This contrasts with the misconception by Kautsky et al. that the method diverges, or the recent discovery by Yang and Tits (1995) that the method converges to stationary points.

5

Artykuł dostępny w postaci pełnego tekstu - kliknij by otworzyć plik

Robust LQR and LQI control with actuator failure of a 2DOF unmanned bicycle robot stabilized by an inertial wheel

100%

Owczarkowski A., Horla D.

International Journal of Applied Mathematics and Computer Science

|

2016

|

tom 26

|

nr 2

325-334

EN

Essential ingredients for robust control are the ability to cope with different types of system behavior following modeling imperfections and the ability to assure a certain performance level. In this paper, we propose to use an actuator fault-tolerant control law to govern, during experiments, the stabilization of a bicycle robot with an inertial wheel in order to take into account unmodeled uncertainty introduced by using a linearized model in an LQR fashion. Our proposal is illustrated by signal plots and the values of performance indices obtained from a set of experiments.

6

Robustness of estimation of first-order autoregressive model under contaminated uniform white noise

88%

Nouali K.

Discussiones Mathematicae Probability and Statistics

|

2009

|

tom 29

|

nr 1

53-68

EN

The first-order autoregressive model with uniform innovations is considered. In this paper, we study the bias-robustness and MSE-robustness of modified maximum likelihood estimator of parameter of the model against departures from distribution of white noise. We used the generalized Beta distribution to describe these departures.

7

Integrated design of observer based fault detection for a class of uncertain nonlinear systems

88%

Chen W., Khan A., Abid M., Ding S.

International Journal of Applied Mathematics and Computer Science

|

2011

|

tom 21

|

nr 3

423-430

EN

Integrated design of observer based Fault Detection (FD) for a class of uncertain nonlinear systems with Lipschitz nonlinearities is studied. In the context of norm based residual evaluation, the residual generator and evaluator are designed together in an integrated form, and, based on it, a trade-off FD system is finally achieved in the sense that, for a given Fault Detection Rate (FDR), the False Alarm Rate (FAR) is minimized. A numerical example is given to illustrate the effectiveness of the proposed design method.

8

Robust fault detection of singular LPV systems with multiple time-varying delays

88%

Hassanabadi A., Shafiee M., Puig V.

International Journal of Applied Mathematics and Computer Science

|

2016

|

tom 26

|

nr 1

45-61

EN

In this paper, the robust fault detection problem for LPV singular delayed systems in the presence of disturbances and actuator faults is considered. For both disturbance decoupling and actuator fault detection, an unknown input observer (UIO) is proposed. The aim is to compute a residual signal which has minimum sensitivity to disturbances while having maximum sensitivity to faults. Robustness to unknown inputs is formulated in the sense of the H∞ -norm by means of the bounded real lemma (BRL) for LPV delayed systems. In order to formulate fault sensitivity conditions, a reference model which characterizes the ideal residual behavior in a faulty situation is considered. The residual error with respect to this reference model is computed. Then, the maximization of the residual fault effect is converted to minimization of its effect on the residual error and is addressed by using the BRL. The compromise between the unknown input effect and the fault effect on the residual is translated into a multi-objective optimization problem with some LMI constraints. In order to show the efficiency and applicability of the proposed method, a part of the Barcelona sewer system is considered.

9

Fault diagnosis and fault tolerant control using set-membership approaches: Application to real case studies

88%

Puig V.

International Journal of Applied Mathematics and Computer Science

|

2010

|

tom 20

|

nr 4

619-635

EN

This paper reviews the use of set-membership methods in fault diagnosis (FD) and fault tolerant control (FTC). Setmembership methods use a deterministic unknown-but-bounded description of noise and parametric uncertainty (interval models). These methods aims at checking the consistency between observed and predicted behaviour by using simple sets to approximate the exact set of possible behaviour (in the parameter or the state space). When an inconsistency is detected between the measured and predicted behaviours obtained using a faultless system model, a fault can be indicated. Otherwise, nothing can be stated. The same principle can be used to identify interval models for fault detection and to develop methods for fault tolerance evaluation. Finally, some real applications will be used to illustrate the usefulness and performance of set-membership methods for FD and FTC.

10

Fault detection and isolation with robust principal component analysis

75%

Tharrault Y., Mourot G., Ragot J., Maquin D.

International Journal of Applied Mathematics and Computer Science

|

2008

|

tom 18

|

nr 4

429-442

EN

Principal component analysis (PCA) is a powerful fault detection and isolation method. However, the classical PCA, which is based on the estimation of the sample mean and covariance matrix of the data, is very sensitive to outliers in the training data set. Usually robust principal component analysis is applied to remove the effect of outliers on the PCA model. In this paper, a fast two-step algorithm is proposed. First, the objective was to find an accurate estimate of the covariance matrix of the data so that a PCA model might be developed that could then be used for fault detection and isolation. A very simple estimate derived from a one-step weighted variance-covariance estimate is used (Ruiz-Gazen, 1996). This is a “local” matrix of variance which tends to emphasize the contribution of close observations in comparison with distant observations (outliers). Second, structured residuals are used for multiple fault detection and isolation. These structured residuals are based on the reconstruction principle, and the existence condition of such residuals is used to determine the detectable faults and the isolable faults. The proposed scheme avoids the combinatorial explosion of faulty scenarios related to multiple faults to be considered. Then, this procedure for outliers detection and isolation is successfully applied to an example with multiple faults.

11

Continuity of solutions of Riccati equations for the discrete-time JLQP

75%

Czornik A., Świerniak A.

International Journal of Applied Mathematics and Computer Science

|

2002

|

tom 12

|

nr 4

539-543

EN

The continuity of the solutions of difference and algebraic coupled Riccati equations for the discrete-time Markovian jump linear quadratic control problem as a function of coefficients is verified. The line of reasoning goes through the use of the minimum property formulated analogously to the one for coupled continuous Riccati equations presented by Wonham and a set of comparison theorems.

Ograniczanie wyników

Towards robustness in neural network based fault diagnosis

Robust multisensor fault tolerant model-following MPC design for constrained systems

Advances in model-based fault diagnosis with evolutionary algorithms and neural networks

Optimization and pole assignment in control system design

Robust LQR and LQI control with actuator failure of a 2DOF unmanned bicycle robot stabilized by an inertial wheel

Robustness of estimation of first-order autoregressive model under contaminated uniform white noise

Integrated design of observer based fault detection for a class of uncertain nonlinear systems

Robust fault detection of singular LPV systems with multiple time-varying delays

Fault diagnosis and fault tolerant control using set-membership approaches: Application to real case studies

Fault detection and isolation with robust principal component analysis

Continuity of solutions of Riccati equations for the discrete-time JLQP