Neural network modeling of proton exchange membrane fuel cell

Sachin V. Puranik, Ali Keyhani, Farshad Khorrami

Research output: Contribution to journalArticle

Abstract

This paper proposes a neural network model of a 500-W proton exchange membrane (PEM) fuel cell. The nonlinear autoregressive moving average model of the PEM fuel cell with external inputs is developed using the recurrent neural networks. The data required to train the neural network model is generated by simulating the nonlinear state space model of the 500-W PEM fuel cell. It is shown that the two-layer neural network, with a hyperbolic tangent sigmoid function, as an activation function, in the first layer, and a pure linear function, as an activation function, in the second layer can effectively model the nonlinear dynamics of the PEM fuel cell. After model is trained and validated, it is used to analyze the dynamic behavior of the PEM fuel cell. Finally, the effect of measurement noise on the performance of the neural network model is investigated, and the results are shown.

Original languageEnglish (US)
Article number5437311
Pages (from-to)474-483
Number of pages10
JournalIEEE Transactions on Energy Conversion
Volume25
Issue number2
DOIs
StatePublished - Jun 2010

Fingerprint

Proton exchange membrane fuel cells (PEMFC)
Neural networks
Chemical activation
Recurrent neural networks

Keywords

  • Fuel cells
  • Modeling
  • Recurrent neural networks

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Energy Engineering and Power Technology

Cite this

Neural network modeling of proton exchange membrane fuel cell. / Puranik, Sachin V.; Keyhani, Ali; Khorrami, Farshad.

In: IEEE Transactions on Energy Conversion, Vol. 25, No. 2, 5437311, 06.2010, p. 474-483.

Research output: Contribution to journalArticle

Puranik, Sachin V. ; Keyhani, Ali ; Khorrami, Farshad. / Neural network modeling of proton exchange membrane fuel cell. In: IEEE Transactions on Energy Conversion. 2010 ; Vol. 25, No. 2. pp. 474-483.
@article{4e25f547c4514bcb81445d53de64db39,
title = "Neural network modeling of proton exchange membrane fuel cell",
abstract = "This paper proposes a neural network model of a 500-W proton exchange membrane (PEM) fuel cell. The nonlinear autoregressive moving average model of the PEM fuel cell with external inputs is developed using the recurrent neural networks. The data required to train the neural network model is generated by simulating the nonlinear state space model of the 500-W PEM fuel cell. It is shown that the two-layer neural network, with a hyperbolic tangent sigmoid function, as an activation function, in the first layer, and a pure linear function, as an activation function, in the second layer can effectively model the nonlinear dynamics of the PEM fuel cell. After model is trained and validated, it is used to analyze the dynamic behavior of the PEM fuel cell. Finally, the effect of measurement noise on the performance of the neural network model is investigated, and the results are shown.",
keywords = "Fuel cells, Modeling, Recurrent neural networks",
author = "Puranik, {Sachin V.} and Ali Keyhani and Farshad Khorrami",
year = "2010",
month = "6",
doi = "10.1109/TEC.2009.2035691",
language = "English (US)",
volume = "25",
pages = "474--483",
journal = "IEEE Transactions on Energy Conversion",
issn = "0885-8969",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "2",

}

TY - JOUR

T1 - Neural network modeling of proton exchange membrane fuel cell

AU - Puranik, Sachin V.

AU - Keyhani, Ali

AU - Khorrami, Farshad

PY - 2010/6

Y1 - 2010/6

N2 - This paper proposes a neural network model of a 500-W proton exchange membrane (PEM) fuel cell. The nonlinear autoregressive moving average model of the PEM fuel cell with external inputs is developed using the recurrent neural networks. The data required to train the neural network model is generated by simulating the nonlinear state space model of the 500-W PEM fuel cell. It is shown that the two-layer neural network, with a hyperbolic tangent sigmoid function, as an activation function, in the first layer, and a pure linear function, as an activation function, in the second layer can effectively model the nonlinear dynamics of the PEM fuel cell. After model is trained and validated, it is used to analyze the dynamic behavior of the PEM fuel cell. Finally, the effect of measurement noise on the performance of the neural network model is investigated, and the results are shown.

AB - This paper proposes a neural network model of a 500-W proton exchange membrane (PEM) fuel cell. The nonlinear autoregressive moving average model of the PEM fuel cell with external inputs is developed using the recurrent neural networks. The data required to train the neural network model is generated by simulating the nonlinear state space model of the 500-W PEM fuel cell. It is shown that the two-layer neural network, with a hyperbolic tangent sigmoid function, as an activation function, in the first layer, and a pure linear function, as an activation function, in the second layer can effectively model the nonlinear dynamics of the PEM fuel cell. After model is trained and validated, it is used to analyze the dynamic behavior of the PEM fuel cell. Finally, the effect of measurement noise on the performance of the neural network model is investigated, and the results are shown.

KW - Fuel cells

KW - Modeling

KW - Recurrent neural networks

UR - http://www.scopus.com/inward/record.url?scp=77952958342&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77952958342&partnerID=8YFLogxK

U2 - 10.1109/TEC.2009.2035691

DO - 10.1109/TEC.2009.2035691

M3 - Article

AN - SCOPUS:77952958342

VL - 25

SP - 474

EP - 483

JO - IEEE Transactions on Energy Conversion

JF - IEEE Transactions on Energy Conversion

SN - 0885-8969

IS - 2

M1 - 5437311

ER -