On the use of neural networks and statistical tools for nonlinear modeling and on-field diagnosis of solid oxide fuel cell stacks

M. Sorrentino (Corresponding Author), D. Marra, C. Pianese, M. Guida, F. Postiglione, K. Wang, Antti Pohjoranta

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

17 Citations (Scopus)

Abstract

The paper reports on the activities performed within the European funded project GENIUS to develop black-box models for modeling and diagnosis of solid oxide fuel cell (SOFC) stacks. Two modeling techniques were investigated, i.e. Neural Networks (NNs) and Statistical Tools (STs). The deployment of NNs was twofold: Recurrent Neural Networks (RNNs) and an NN classifier were developed to simulate transient operation of SOFCs and identify some specific faults that may occur in such devices, respectively. On the other hand, STs are based on a stepwise multiple regression. Data for model development were obtained from experiments specifically designed to reach maximal information content. The final aim was to obtain highly general models of SOFC stacks' operation in both transient and steady state. All the developed black-box models exhibited high accuracy and reliability on both training and test data-sets. Moreover, the black-box models were also proven effective in performing real-time monitoring and degradation analysis for different SOFC stack technologies.
Original languageEnglish
Pages (from-to)298-307
Number of pages9
JournalEnergy Procedia
Volume45
DOIs
Publication statusPublished - 2013
MoE publication typeA4 Article in a conference publication
Event68th Conference of the Italian Thermal Machines Engineering Association, ATI 2013 - Bologna, Italy
Duration: 11 Sep 201313 Sep 2013
Conference number: 68

Fingerprint

Solid oxide fuel cells (SOFC)
Neural networks
Recurrent neural networks
Classifiers
Degradation
Monitoring
Experiments

Keywords

  • Black-box models
  • neural network based classification
  • recurrent neural network
  • solid oxide fuel cells
  • step-wise regression analysis

Cite this

Sorrentino, M. ; Marra, D. ; Pianese, C. ; Guida, M. ; Postiglione, F. ; Wang, K. ; Pohjoranta, Antti. / On the use of neural networks and statistical tools for nonlinear modeling and on-field diagnosis of solid oxide fuel cell stacks. In: Energy Procedia. 2013 ; Vol. 45. pp. 298-307.
@article{c4c8898b009c4f4c80e45f83e03ab8a1,
title = "On the use of neural networks and statistical tools for nonlinear modeling and on-field diagnosis of solid oxide fuel cell stacks",
abstract = "The paper reports on the activities performed within the European funded project GENIUS to develop black-box models for modeling and diagnosis of solid oxide fuel cell (SOFC) stacks. Two modeling techniques were investigated, i.e. Neural Networks (NNs) and Statistical Tools (STs). The deployment of NNs was twofold: Recurrent Neural Networks (RNNs) and an NN classifier were developed to simulate transient operation of SOFCs and identify some specific faults that may occur in such devices, respectively. On the other hand, STs are based on a stepwise multiple regression. Data for model development were obtained from experiments specifically designed to reach maximal information content. The final aim was to obtain highly general models of SOFC stacks' operation in both transient and steady state. All the developed black-box models exhibited high accuracy and reliability on both training and test data-sets. Moreover, the black-box models were also proven effective in performing real-time monitoring and degradation analysis for different SOFC stack technologies.",
keywords = "Black-box models, neural network based classification, recurrent neural network, solid oxide fuel cells, step-wise regression analysis",
author = "M. Sorrentino and D. Marra and C. Pianese and M. Guida and F. Postiglione and K. Wang and Antti Pohjoranta",
note = "CA2: BA2154",
year = "2013",
doi = "10.1016/j.egypro.2014.01.032",
language = "English",
volume = "45",
pages = "298--307",
journal = "Energy Procedia",
issn = "1876-6102",
publisher = "Elsevier",

}

On the use of neural networks and statistical tools for nonlinear modeling and on-field diagnosis of solid oxide fuel cell stacks. / Sorrentino, M. (Corresponding Author); Marra, D.; Pianese, C.; Guida, M.; Postiglione, F.; Wang, K.; Pohjoranta, Antti.

In: Energy Procedia, Vol. 45, 2013, p. 298-307.

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

TY - JOUR

T1 - On the use of neural networks and statistical tools for nonlinear modeling and on-field diagnosis of solid oxide fuel cell stacks

AU - Sorrentino, M.

AU - Marra, D.

AU - Pianese, C.

AU - Guida, M.

AU - Postiglione, F.

AU - Wang, K.

AU - Pohjoranta, Antti

N1 - CA2: BA2154

PY - 2013

Y1 - 2013

N2 - The paper reports on the activities performed within the European funded project GENIUS to develop black-box models for modeling and diagnosis of solid oxide fuel cell (SOFC) stacks. Two modeling techniques were investigated, i.e. Neural Networks (NNs) and Statistical Tools (STs). The deployment of NNs was twofold: Recurrent Neural Networks (RNNs) and an NN classifier were developed to simulate transient operation of SOFCs and identify some specific faults that may occur in such devices, respectively. On the other hand, STs are based on a stepwise multiple regression. Data for model development were obtained from experiments specifically designed to reach maximal information content. The final aim was to obtain highly general models of SOFC stacks' operation in both transient and steady state. All the developed black-box models exhibited high accuracy and reliability on both training and test data-sets. Moreover, the black-box models were also proven effective in performing real-time monitoring and degradation analysis for different SOFC stack technologies.

AB - The paper reports on the activities performed within the European funded project GENIUS to develop black-box models for modeling and diagnosis of solid oxide fuel cell (SOFC) stacks. Two modeling techniques were investigated, i.e. Neural Networks (NNs) and Statistical Tools (STs). The deployment of NNs was twofold: Recurrent Neural Networks (RNNs) and an NN classifier were developed to simulate transient operation of SOFCs and identify some specific faults that may occur in such devices, respectively. On the other hand, STs are based on a stepwise multiple regression. Data for model development were obtained from experiments specifically designed to reach maximal information content. The final aim was to obtain highly general models of SOFC stacks' operation in both transient and steady state. All the developed black-box models exhibited high accuracy and reliability on both training and test data-sets. Moreover, the black-box models were also proven effective in performing real-time monitoring and degradation analysis for different SOFC stack technologies.

KW - Black-box models

KW - neural network based classification

KW - recurrent neural network

KW - solid oxide fuel cells

KW - step-wise regression analysis

U2 - 10.1016/j.egypro.2014.01.032

DO - 10.1016/j.egypro.2014.01.032

M3 - Article in a proceedings journal

VL - 45

SP - 298

EP - 307

JO - Energy Procedia

JF - Energy Procedia

SN - 1876-6102

ER -