Treffer: Addressing Data Scarcity in PEMFC Fault Diagnosis Using Adversarial Learning

Title:
Addressing Data Scarcity in PEMFC Fault Diagnosis Using Adversarial Learning
Contributors:
Laboratoire d'Innovation pour les Technologies des Energies Nouvelles et les nanomatériaux (LITEN / CEA-DES), CEA-Direction des Energies (ex-Direction de l'Energie Nucléaire) (CEA-DES (ex-DEN)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Institut National de L'Energie Solaire (INES), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Savoie Mont Blanc (USMB [Université de Savoie] [Université de Chambéry])-Centre National de la Recherche Scientifique (CNRS)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Université Savoie Mont Blanc (USMB [Université de Savoie] [Université de Chambéry])-Centre National de la Recherche Scientifique (CNRS), Commissariat à l'énergie atomique et aux énergies alternatives - Laboratoire d'Electronique et de Technologie de l'Information (CEA-LETI), Direction de Recherche Technologique (CEA) (DRT (CEA)), Commissariat à l'énergie atomique et aux énergies alternatives (CEA)-Commissariat à l'énergie atomique et aux énergies alternatives (CEA), PEPRH2, national French project, called Durasys-PAC (https://www.pepr-hydrogene.fr/projets/durasys-pac/)
Source:
2024 IEEE International Conference on Prognostics and Health Management (ICPHM). :393-398
Publisher Information:
HAL CCSD; IEEE, 2024.
Publication Year:
2024
Collection:
collection:CEA
collection:UNIV-SAVOIE
collection:CNRS
collection:DRT
collection:DEN
collection:TDS-MACS
collection:LETI
collection:LITEN
collection:CEA-GRE
collection:INES
Subject Geographic:
Original Identifier:
HAL:
Document Type:
Konferenz conferenceObject<br />Conference papers
Language:
English
Relation:
info:eu-repo/semantics/altIdentifier/doi/10.1109/ICPHM61352.2024.10627464
DOI:
10.1109/ICPHM61352.2024.10627464
Accession Number:
edshal.cea.04760251v1
Database:
HAL

Weitere Informationen

To pave the way to a large industrial deployment of proton exchange membrane fuel cells (PEMFCs) it is essential to improve the durability of the technology. Achieving this requires accurate diagnosis of abnormal operating conditions. To automate the monitoring process in PEMFC, data-driven fault diagnosis models have shown great potential. However, in practice, the performance of data-driven models can be compromised by the availability of the data for training these algorithms. This article proposes a simulation-driven domain adaptation method to circumvent the data scarcity issue using a physics-based model. Through adversarial training, we focus on extracting relevant, high level features for the fault diagnosis task from the simulated dataset, while simultaneously aligning the features extracted from a real dataset acquired on a PEMFC stack. Introducing supervision on the real dataset, we are able to extract features that are discriminative for the fault diagnosis task as well as invariant to the domain. The experimental results validate the effectiveness of the proposed method in dealing with scenarios of insufficient data availability for PEMFC fault diagnosis.