2018 |
Houiji, Marwa; Hamdaoui, Rim; Aoun, Mohamed 2018, (Cited by: 1). Résumé | Liens | BibTeX | Étiquettes: Actuator and sensor faults, Actuators, Adaptive threshold adjustments, Adaptive thresholds, ARThSK, Failure analysis, Fault accommodation, Fault detection, Linear stochastic system, Robust fault detection, Robust fault diagnosis, Stochastic control systems, Stochastic systems @conference{Houiji2018810b, This paper presents the problem of robust fault diagnosis and accommodation for a class of linear stochastic systems where simultaneous actuator and sensor faults may occur at a given time. Firstly, based on Augmented Robust Three stage Kalman filters (ARThSKF) we obtained both the fault estimation and the residual signal. Then, residual evaluation is achieved by making use of an adaptive threshold adjustment algorithm based on the grey theory. Obtained results show that the false-alarm rates and the missing alarm rates are minimized by the developed method; also this approach detects small or incipient faults more effectively than the classical robust fault detection algorithms with fixed thresholds. Finally, an additive control input is introduced for cancelling out the fault’s effect on the system. We evaluate our proposal through simulation and we demonstrate its feasibility. © 2018 IEEE. |
Houiji, Marwa; Hamdaoui, Rim; Aoun, Mohamed 2018, (Cited by: 1). Résumé | Liens | BibTeX | Étiquettes: Actuator and sensor faults, Actuators, Adaptive threshold adjustments, Adaptive thresholds, ARThSK, Failure analysis, Fault accommodation, Fault detection, Linear stochastic system, Robust fault detection, Robust fault diagnosis, Stochastic control systems, Stochastic systems @conference{Houiji2018810, This paper presents the problem of robust fault diagnosis and accommodation for a class of linear stochastic systems where simultaneous actuator and sensor faults may occur at a given time. Firstly, based on Augmented Robust Three stage Kalman filters (ARThSKF) we obtained both the fault estimation and the residual signal. Then, residual evaluation is achieved by making use of an adaptive threshold adjustment algorithm based on the grey theory. Obtained results show that the false-alarm rates and the missing alarm rates are minimized by the developed method; also this approach detects small or incipient faults more effectively than the classical robust fault detection algorithms with fixed thresholds. Finally, an additive control input is introduced for cancelling out the fault’s effect on the system. We evaluate our proposal through simulation and we demonstrate its feasibility. © 2018 IEEE. |
Publications
2018 |
2018, (Cited by: 1). |
2018, (Cited by: 1). |