TY - JOUR
T1 - Design and application of fault tolerant control for multi-sensor fault of nonlinear multi-variables system
AU - Bo, Cuimei
AU - Wang, Zhiquan
AU - Wang, Yonghua
AU - Zhang, Shi
AU - Lin, Jinguo
PY - 2006/7
Y1 - 2006/7
N2 - Aiming at the sensor fault of multi-variable nonlinear system, a novel fault tolerant control (FTC) method based on neural network models is investigated. The method colligates weighting moving average residual within L time window to detect multi-type sensor fault, and uses neural network technique to online re-construct sensor signal, and achieves the tolerant proportional integral derivative (PID) control. The three-tank with multiple sensor fault concurrence is simulated, the simulation result shows that the fault detection and tolerant control strategy proposed have stronger robustness and tolerant fault ability.
AB - Aiming at the sensor fault of multi-variable nonlinear system, a novel fault tolerant control (FTC) method based on neural network models is investigated. The method colligates weighting moving average residual within L time window to detect multi-type sensor fault, and uses neural network technique to online re-construct sensor signal, and achieves the tolerant proportional integral derivative (PID) control. The three-tank with multiple sensor fault concurrence is simulated, the simulation result shows that the fault detection and tolerant control strategy proposed have stronger robustness and tolerant fault ability.
KW - Fault detection and diagnosis
KW - Fault-tolerant PID control
KW - Multi-variables system
KW - Neural network model
KW - Three-tank process
UR - http://www.scopus.com/inward/record.url?scp=33750212771&partnerID=8YFLogxK
M3 - 文章
AN - SCOPUS:33750212771
SN - 1001-0505
VL - 36
SP - 43
EP - 47
JO - Dongnan Daxue Xuebao (Ziran Kexue Ban)/Journal of Southeast University (Natural Science Edition)
JF - Dongnan Daxue Xuebao (Ziran Kexue Ban)/Journal of Southeast University (Natural Science Edition)
IS - SUPPL.
ER -