Geodesic
Neural network-based fault diagnosis and fault-tolerant control for nonlinear systems with output measurement noise
Kybernetika, Tome 60 (2024) no. 2, pp. 244-270 Cet article a éte moissonné depuis la source Czech Digital Mathematics Library

Voir la notice de l'article

In this article, the problems of fault diagnosis (FD) and fault-tolerant control (FTC) are investigated for a class of nonlinear systems with output measurement noise. Due to the influence of measurement noise in the output sensor, the output observation error cannot be accurately obtained, which causes obstacles to the accuracy of FD. To address this issue, an output filter and disturbance estimator are constructed to decrease the negative effects of measurement noise and observer gain disturbances, and a novel non-fragile neural observer is designed to estimate the unknown states. A new evaluation function is also introduced to detect faults. Then, a novel neural FTC controller is proposed in the presence of faults, to ensure that all the closed-loop system signals are semiglobally uniformly ultimately bounded (SGUUB). The effectiveness of the proposed methodology is verified via numerical simulation of a one-link robot system.
In this article, the problems of fault diagnosis (FD) and fault-tolerant control (FTC) are investigated for a class of nonlinear systems with output measurement noise. Due to the influence of measurement noise in the output sensor, the output observation error cannot be accurately obtained, which causes obstacles to the accuracy of FD. To address this issue, an output filter and disturbance estimator are constructed to decrease the negative effects of measurement noise and observer gain disturbances, and a novel non-fragile neural observer is designed to estimate the unknown states. A new evaluation function is also introduced to detect faults. Then, a novel neural FTC controller is proposed in the presence of faults, to ensure that all the closed-loop system signals are semiglobally uniformly ultimately bounded (SGUUB). The effectiveness of the proposed methodology is verified via numerical simulation of a one-link robot system.
DOI : 10.14736/kyb-2024-2-0244
Classification : 93C10, 94C12
Keywords: fault diagnosis; fault-tolerant control; output measurement noise; non-fragile; output filter 
@article{10_14736_kyb_2024_2_0244,
     author = {Shen, Yanjun and Ma, Chen and Zhao, Chenhao and Wu, Zebin},
     title = {Neural network-based fault diagnosis and fault-tolerant control for nonlinear systems with output measurement noise},
     journal = {Kybernetika},
     pages = {244--270},
     year = {2024},
     volume = {60},
     number = {2},
     doi = {10.14736/kyb-2024-2-0244},
     mrnumber = {4757772},
     zbl = {07893457},
     language = {en},
     url = {http://geodesic.mathdoc.fr/articles/10.14736/kyb-2024-2-0244/}
}
TY  - JOUR
AU  - Shen, Yanjun
AU  - Ma, Chen
AU  - Zhao, Chenhao
AU  - Wu, Zebin
TI  - Neural network-based fault diagnosis and fault-tolerant control for nonlinear systems with output measurement noise
JO  - Kybernetika
PY  - 2024
SP  - 244
EP  - 270
VL  - 60
IS  - 2
UR  - http://geodesic.mathdoc.fr/articles/10.14736/kyb-2024-2-0244/
DO  - 10.14736/kyb-2024-2-0244
LA  - en
ID  - 10_14736_kyb_2024_2_0244
ER  - 
%0 Journal Article
%A Shen, Yanjun
%A Ma, Chen
%A Zhao, Chenhao
%A Wu, Zebin
%T Neural network-based fault diagnosis and fault-tolerant control for nonlinear systems with output measurement noise
%J Kybernetika
%D 2024
%P 244-270
%V 60
%N 2
%U http://geodesic.mathdoc.fr/articles/10.14736/kyb-2024-2-0244/
%R 10.14736/kyb-2024-2-0244
%G en
%F 10_14736_kyb_2024_2_0244
Shen, Yanjun; Ma, Chen; Zhao, Chenhao; Wu, Zebin. Neural network-based fault diagnosis and fault-tolerant control for nonlinear systems with output measurement noise. Kybernetika, Tome 60 (2024) no. 2, pp. 244-270. doi: 10.14736/kyb-2024-2-0244

Cité par Sources :