Model-free fault detection: application to Polymer Electrolyte Fuel Cell system - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Model-free fault detection: application to Polymer Electrolyte Fuel Cell system

Résumé

A novel fault detection method is presented in this paper. The proposed method can be considered as an extension of the model-free control which is based on an ultra-local model. The model-free controller has demonstrated a great ability to perform the control tasks despite the presence of a fault. This undeniable advantage can mask the effect of a fault resulting in an abnormal degradation impacting one of the components ensuring the control loop or the system itself. The proposed method allows coping with this type of event rapid coping. The basic idea of the diagnostic method is to reconstruct the measured output of the system from the ultra-local model used for system control. The estimated output is compared with the measured one to generate a residual which is employed as the fault indicator. An experimental validation of the polymer electrolyte membrane fuel cell (PEMFC) system is carried out in real time to detect a a system fault that results in membrane flooding. The main advantage of the method is to be able to simultaneously control the system and detect faults that may affect it without needing an accurate knowledge of the mathematical/physical system model. The results obtained are very promising for real-time diagnosis of the polymer electrolyte membrane fuel cell systems.
Fichier principal
Vignette du fichier
6391b61b-d6d7-4824-9815-5f7d314afbca-author.pdf (2.48 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04172754 , version 1 (28-07-2023)

Identifiants

Citer

Meziane Ait Ziane, Nadia Yousfi Steiner, Cédric Join, Michel Benne, Cédric Damour, et al.. Model-free fault detection: application to Polymer Electrolyte Fuel Cell system. International Conference on Systems and Control, ICSC 2022, Nov 2022, Marseille, France. ⟨10.1109/ICSC57768.2022.9993863⟩. ⟨hal-04172754⟩
42 Consultations
56 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More