Artificial Neural Network Modeling of Pt/C Cathode Degradation in PEM Fuel Cells

被引:0
作者
Erfan Maleki
Nasim Maleki
机构
[1] Sharif University of Technology,Department of Mechanical Engineering
[2] Razi University,Department of Chemistry
来源
Journal of Electronic Materials | 2016年 / 45卷
关键词
PEM fuel cell; cathode degradation; modeling; artificial neural network;
D O I
暂无
中图分类号
学科分类号
摘要
Use of computational modeling with a few experiments is considered useful to obtain the best possible result for a final product, without performing expensive and time-consuming experiments. Proton exchange membrane fuel cells (PEMFCs) can produce clean electricity, but still require further study. An oxygen reduction reaction (ORR) takes place at the cathode, and carbon-supported platinum (Pt/C) is commonly used as an electrocatalyst. The harsh conditions during PEMFC operation result in Pt/C degradation. Observation of changes in the Pt/C layer under operating conditions provides a tool to study the lifetime of PEMFCs and overcome durability issues. Recently, artificial neural networks (ANNs) have been used to solve, predict, and optimize a wide range of scientific problems. In this study, several rates of change at the cathode were modeled using ANNs. The backpropagation (BP) algorithm was used to train the network, and experimental data were employed for network training and testing. Two different models are constructed in the present study. First, the potential cycles, temperature, and humidity are used as inputs to predict the resulting Pt dissolution rate of the Pt/C at the cathode as the output parameter of the network. Thereafter, the Pt dissolution rate and Pt ion diffusivity are regarded as inputs to obtain values of the Pt particle radius change rate, Pt mass loss rate, and surface area loss rate as outputs. The networks are finely tuned, and the modeling results agree well with experimental data. The modeled responses of the ANNs are acceptable for this application.
引用
收藏
页码:3822 / 3834
页数:12
相关论文
共 50 条
[21]   Artificial Neural Network Modeling in Pretreatment of Garden Biomass for Lignocellulose Degradation [J].
Bhange, Vivek P. ;
Bhivgade, Urvashi V. ;
Vaidya, Atul N. .
WASTE AND BIOMASS VALORIZATION, 2019, 10 (06) :1571-1583
[22]   The effect of Nafion film on the cathode catalyst layer performance in a low-Pt PEM fuel cell [J].
Kulikovsky, Andrei .
ELECTROCHEMISTRY COMMUNICATIONS, 2019, 103 :61-65
[23]   Modeling of fuel consumption using artificial neural networks [J].
WITASZEK K. .
Diagnostyka, 2020, 21 (04) :103-113
[24]   Neural network model for a commercial PEM fuel cell system [J].
Saenrung, Anucha ;
Abtahi, Amir ;
Zilouchian, Ali .
JOURNAL OF POWER SOURCES, 2007, 172 (02) :749-759
[25]   Modeling for Optimal Operation of PEM Fuel Cells and Electrolyzers [J].
Gabrielli, Paolo ;
Flamm, Ben ;
Eichler, Annika ;
Gazzani, Matteo ;
Lygeros, John ;
Mazzotti, Marco .
2016 IEEE 16TH INTERNATIONAL CONFERENCE ON ENVIRONMENT AND ELECTRICAL ENGINEERING (EEEIC), 2016,
[26]   Computational modeling of PEM fuel cells with PBI membranes [J].
Cheddie, Denver F. ;
Munroe, Norman D. H. .
PROCEEDINGS OF THE 4TH INTERNATIONAL CONFERENCE ON FUEL CELL SCIENCE, ENGINEERING, AND TECHNOLOGY, PTS A AND B, 2006, :243-252
[27]   Mathematical modeling of water management in PEM fuel cells [J].
Berg, P ;
Promislow, K ;
Stockie, J ;
Wetton, B .
NANOTECH 2003, VOL 3, 2003, :459-462
[28]   Modeling and Optimization of Anode-Supported Solid Oxide Fuel Cells on Cell Parameters via Artificial Neural Network and Genetic Algorithm [J].
Bozorgmehri, S. ;
Hamedi, M. .
FUEL CELLS, 2012, 12 (01) :11-23
[29]   Transient analysis for the cathode gas diffusion layer of PEM fuel cells [J].
Song, Datong ;
Wang, Qianpu ;
Liu, Zhong-Sheng ;
Huang, Cheng .
JOURNAL OF POWER SOURCES, 2006, 159 (02) :928-942
[30]   Stability of platinum based alloy cathode catalysts in PEM fuel cells [J].
Colón-Mercado, HR ;
Popov, BN .
JOURNAL OF POWER SOURCES, 2006, 155 (02) :253-263