Particle Filter Based Prognostics of PEM ...
Document type :
Communication dans un congrès avec actes
Title :
Particle Filter Based Prognostics of PEM Fuel Cell in Bond Graph Framework
Author(s) :
Jha, Mayank Shekhar [Auteur]
Méthodes et Outils pour la Conception Intégrée de Systèmes [MOCIS]
Bressel, Mathieu [Auteur]
Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174) [FEMTO-ST]
Bouamama, Belkacem [Auteur]
Méthodes et Outils pour la Conception Intégrée de Systèmes [MOCIS]
Tanguy, Geneviève [Auteur]
Méthodes et Outils pour la Conception Intégrée de Systèmes [MOCIS]
Hilairet, Mickaël [Auteur]
Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174) [FEMTO-ST]
Hissel, Daniel [Auteur]
Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174) [FEMTO-ST]
Méthodes et Outils pour la Conception Intégrée de Systèmes [MOCIS]
Bressel, Mathieu [Auteur]
Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174) [FEMTO-ST]
Bouamama, Belkacem [Auteur]
Méthodes et Outils pour la Conception Intégrée de Systèmes [MOCIS]
Tanguy, Geneviève [Auteur]
Méthodes et Outils pour la Conception Intégrée de Systèmes [MOCIS]
Hilairet, Mickaël [Auteur]
Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174) [FEMTO-ST]
Hissel, Daniel [Auteur]
Franche-Comté Électronique Mécanique, Thermique et Optique - Sciences et Technologies (UMR 6174) [FEMTO-ST]
Conference title :
Conférence Internationale des Energies Renouvelables
City :
Sousse
Country :
Tunisie
Start date of the conference :
2015-12-21
HAL domain(s) :
Physique [physics]/Mécanique [physics]/Mécanique des fluides [physics.class-ph]
Physique [physics]/Mécanique [physics]/Thermique [physics.class-ph]
Sciences de l'ingénieur [physics]/Automatique / Robotique
Sciences de l'ingénieur [physics]/Energie électrique
Physique [physics]/Mécanique [physics]/Thermique [physics.class-ph]
Sciences de l'ingénieur [physics]/Automatique / Robotique
Sciences de l'ingénieur [physics]/Energie électrique
English abstract : [en]
This paper develops an efficient solution towards the prognostics of industrial PEMFC. It involves an efficient multi-energetic model suited for diagnostics and prognostics, developed in Bond Graph framework. The benefits ...
Show more >This paper develops an efficient solution towards the prognostics of industrial PEMFC. It involves an efficient multi-energetic model suited for diagnostics and prognostics, developed in Bond Graph framework. The benefits of Particle Filters (PF) is integrated with the BG model derived Analytical Redundancy Relations (ARRs), for prognostics of the electrical-electrochemical (EE) part. The prognostic problem is treated as the joint state-parameter estimation problem in Particle Filter framework, a hybrid prognostic approach wherein, a fault model is constructed in state-space. The state equation is inspired from the statistical degradation model of the global resistance and limiting current. Observation equation is obtained from the Analytical Redundancy Relations (ARRs) derived from BG model. Using PF algorithms, estimation of SOH is obtained along with the estimation of the associated hidden time-varying parameters that influence the progression of degradation. The latter is tracked to obtain the SOH in probabilistic terms. This in turn is used for prediction of Remaining Useful Life of the EE part of PEMFC. The methodology is applied on real degradation data sets under constant load current profile.Show less >
Show more >This paper develops an efficient solution towards the prognostics of industrial PEMFC. It involves an efficient multi-energetic model suited for diagnostics and prognostics, developed in Bond Graph framework. The benefits of Particle Filters (PF) is integrated with the BG model derived Analytical Redundancy Relations (ARRs), for prognostics of the electrical-electrochemical (EE) part. The prognostic problem is treated as the joint state-parameter estimation problem in Particle Filter framework, a hybrid prognostic approach wherein, a fault model is constructed in state-space. The state equation is inspired from the statistical degradation model of the global resistance and limiting current. Observation equation is obtained from the Analytical Redundancy Relations (ARRs) derived from BG model. Using PF algorithms, estimation of SOH is obtained along with the estimation of the associated hidden time-varying parameters that influence the progression of degradation. The latter is tracked to obtain the SOH in probabilistic terms. This in turn is used for prediction of Remaining Useful Life of the EE part of PEMFC. The methodology is applied on real degradation data sets under constant load current profile.Show less >
Language :
Anglais
Peer reviewed article :
Oui
Audience :
Internationale
Popular science :
Non
Collections :
Source :
Files
- https://hal.archives-ouvertes.fr/hal-02868285/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-02868285/document
- Open access
- Access the document
- https://hal.archives-ouvertes.fr/hal-02868285/document
- Open access
- Access the document
- document
- Open access
- Access the document
- 1f7ff03d-67ae-49e5-9622-c64caec0fdaf-author.pdf
- Open access
- Access the document