Fault diagnosis of PEMFC using semi-empirical signal processing and K-means clustering - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Fault diagnosis of PEMFC using semi-empirical signal processing and K-means clustering

Résumé

Research in Proton Exchange Membrane Fuel Cell (PEMFC) systems has received much attention in recent years due to the rapid growth in the use of PEMFC systems across a wide range of applications. Yet they still need to be further enhanced in terms of reliability in order to remain competitive in the market with other established technologies. The present study focuses on the development of a reliable fault identification tool for PEMFCs. This diagnostic tool is based on two steps: firstly, a semi-empirical signal processing approach that allows better extraction of signal information than the traditional methods used; secondly, the design of a K-means clustering algorithm that is trained using a dataset with two separate faults of flooding and drying. The results show a good recognition rate of 95% for the considered faults.
Fichier non déposé

Dates et versions

hal-04452049 , version 1 (12-02-2024)

Identifiants

  • HAL Id : hal-04452049 , version 1

Citer

Abderazek Cheikh, Nadia Yousfi Steiner, Elodie Pahon, Cédric Damour, Michel Benne, et al.. Fault diagnosis of PEMFC using semi-empirical signal processing and K-means clustering. Symposium de Génie Electrique, Jul 2023, Lille, France. ⟨hal-04452049⟩
10 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More