Mechanics & Industry
Volume 17, Number 1, 2016
|Number of page(s)||11|
|Published online||16 October 2015|
Modeling and experimental verification of a 25W fabricated PEM fuel cell by parametric and GMDH-type neural network
Department of Energy, Material and Energy Research Center,
2 Department of Mechanical Engineering, Pardis Branch, Islamic Azad University, Pardis New City, Tehran, Iran
a Corresponding author: firstname.lastname@example.org
Received: 29 November 2014
Accepted: 19 June 2015
In this paper two artificial intelligence techniques to predict and control behavior of a 25W fabricated proton exchange membrane (PEM) fuel cell, have been investigated. These approaches are: “Parametric Neural Network (PNN)” and “Group Method of Data Handling (GMDH)” for the first time. A PNN model is developed by introducing a “p” parameter in the activation function of the neural network. PNN model with its specific tangent hyperbolic transfer function have the ability to be with different nonlinearity degrees of input data. To develop GMDH network, quadratic polynomial was utilized. To determine proper weights of GMDH network, back propagation algorithm has been used. The input layer consists of gas pressure, fuel cell temperature and input current experimental data, to predict the output voltage. The results show that both generalized Parametric and GMDH-type neural networks are reliable tools to predict the output voltage of PEM fuel cell with high coefficient of determination values of 0.96 and 0.98.
Key words: Proton exchange membrane fuel cell / hydrogen modeling / output voltage prediction / parametric neural network / group method of data handling
© AFM, EDP Sciences 2015
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.