SOFC modelling based on discrete Bayesian network for system diagnosis use

K. Wang, M.C. Pera, D. Hissel, N. Yousfi Steiner, Antti Pohjoranta, S. Pofahl

Research output: Contribution to journalArticle in a proceedings journalScientificpeer-review

1 Citation (Scopus)

Abstract

We propose in this paper a diagnosis method that is aimed to detect and isolate SOFC system fault by using the FC stack as a sensor. A discrete Bayesian network (BN) was established to illustrate the input-output causal relations of the stack. In order to examine the generalizability of the network structure, the BN was parameterized to fit the experimental data from two different SOFC systems. The models showed reasonable accuracy of state estimation for 6 operating variables. Finally, the BN model was experimented for diagnosing a specified system fault.
Original languageEnglish
Pages (from-to)675-680
JournalIFAC Proceedings Volumes
Volume45
Issue number21
DOIs
Publication statusPublished - 2012
MoE publication typeA4 Article in a conference publication
Event8th Power Plant and Power System Control Symposium, PPPSC 2012 - Toulouse, France
Duration: 2 Sep 20125 Sep 2012

Fingerprint

Bayesian networks
Solid oxide fuel cells (SOFC)
State estimation
Sensors

Keywords

  • Bayesian network
  • conditional probability
  • fault diagnosis
  • maximum likelihood
  • SOFC

Cite this

Wang, K. ; Pera, M.C. ; Hissel, D. ; Yousfi Steiner, N. ; Pohjoranta, Antti ; Pofahl, S. / SOFC modelling based on discrete Bayesian network for system diagnosis use. In: IFAC Proceedings Volumes. 2012 ; Vol. 45, No. 21. pp. 675-680.
@article{a24f6fa448454120863e9207ecbd0f23,
title = "SOFC modelling based on discrete Bayesian network for system diagnosis use",
abstract = "We propose in this paper a diagnosis method that is aimed to detect and isolate SOFC system fault by using the FC stack as a sensor. A discrete Bayesian network (BN) was established to illustrate the input-output causal relations of the stack. In order to examine the generalizability of the network structure, the BN was parameterized to fit the experimental data from two different SOFC systems. The models showed reasonable accuracy of state estimation for 6 operating variables. Finally, the BN model was experimented for diagnosing a specified system fault.",
keywords = "Bayesian network, conditional probability, fault diagnosis, maximum likelihood, SOFC",
author = "K. Wang and M.C. Pera and D. Hissel and {Yousfi Steiner}, N. and Antti Pohjoranta and S. Pofahl",
year = "2012",
doi = "10.3182/20120902-4-FR-2032.00118",
language = "English",
volume = "45",
pages = "675--680",
journal = "IFAC-PapersOnLine",
issn = "2405-8971",
publisher = "IFAC Secretariat",
number = "21",

}

SOFC modelling based on discrete Bayesian network for system diagnosis use. / Wang, K.; Pera, M.C.; Hissel, D.; Yousfi Steiner, N.; Pohjoranta, Antti; Pofahl, S.

In: IFAC Proceedings Volumes, Vol. 45, No. 21, 2012, p. 675-680.

Research output: Contribution to journalArticle in a proceedings journalScientificpeer-review

TY - JOUR

T1 - SOFC modelling based on discrete Bayesian network for system diagnosis use

AU - Wang, K.

AU - Pera, M.C.

AU - Hissel, D.

AU - Yousfi Steiner, N.

AU - Pohjoranta, Antti

AU - Pofahl, S.

PY - 2012

Y1 - 2012

N2 - We propose in this paper a diagnosis method that is aimed to detect and isolate SOFC system fault by using the FC stack as a sensor. A discrete Bayesian network (BN) was established to illustrate the input-output causal relations of the stack. In order to examine the generalizability of the network structure, the BN was parameterized to fit the experimental data from two different SOFC systems. The models showed reasonable accuracy of state estimation for 6 operating variables. Finally, the BN model was experimented for diagnosing a specified system fault.

AB - We propose in this paper a diagnosis method that is aimed to detect and isolate SOFC system fault by using the FC stack as a sensor. A discrete Bayesian network (BN) was established to illustrate the input-output causal relations of the stack. In order to examine the generalizability of the network structure, the BN was parameterized to fit the experimental data from two different SOFC systems. The models showed reasonable accuracy of state estimation for 6 operating variables. Finally, the BN model was experimented for diagnosing a specified system fault.

KW - Bayesian network

KW - conditional probability

KW - fault diagnosis

KW - maximum likelihood

KW - SOFC

U2 - 10.3182/20120902-4-FR-2032.00118

DO - 10.3182/20120902-4-FR-2032.00118

M3 - Article in a proceedings journal

VL - 45

SP - 675

EP - 680

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

SN - 2405-8971

IS - 21

ER -