Hydrogen production by electrolysis is emerging as one of the most promising ways to meet future fuel demand; likewise, the development of models capable of simulating the operation of electrolysis devices is indispensable in the efficient design of power generation systems, reducing manufacturing costs and resources savings. The nonlinear nature of the Artificial Neural Network (ANN) plays a key role at the development of models for predicting the performance of complex systems. The behavior of a Polymer Electrolyte Membrane (PEM) Electrolyzer of three cell stack (100 cm2 of active area) was modeled successfully using a Multilayer Perceptron Network (MLP). This dynamic model has been trained to learn the internal relationships that govern this electrolysis device and predict its behavior without any physical equations. The electric current supply and the operation temperature were used as input vector able to predict each cell voltage behavior. A reliable accuracy (< 2%) was reached in this work after comparing the single cell performance of the real electrolyzer versus the ANN based model. This predictive model can be used as a virtual device into a more complex energy system.
Artificial Neural Network (ANN), Model, Multilayer Perceptron Network (MLP), PEM Electrolyzer
 J. Ledesma-Garcia, R. Barbosa, T.W. Chapman, L.G. Arriaga, L.A. Godinez, Int. J. Hydrogen Energy, 34, 2008 (2009).
 V.A. Goltsov, T.N. Veziroglu and L.F. Goltsova, Int. J. Hydrogen Energy, 31, 153 (2006).
 C.J. Winter. Int. J. Hydrogen Energy, 30, 681 (2005).
 S.A. Sherif, F. Barbir and T.N. Veziroglu, Sol. Energy, 78, 647 (2005).
 S.A. Sherif, F. Barbir and T.N. Veziroglu, Electr. J., 18, 62 (2005).
 L.G. Arriaga, W. Martinez, U. Cano, H. Blud, Int. J. Hydrogen Energy, 32, 2247 (2007).
 S.A. Grigoriev, V.I. Porembsky, V.N. Fateev, Int. J. Hydrogen Energy, 31, 171 (2006).
 S.D. Greenway, E.B. Fox, A.A. Ekechukwu. Int. J. of Hydrogen Energy, 34, 6603 (2009).
 E. Slavcheva, I. Radev, S. Bliznakov, G. Topalov, P. Andreev, E. Budevski, Electrochem. Acta, 52, 3889 (2007).
 S. Siracusano, V. Baglio, A. Di Blasi, N. Briguglio, A. Stassi, R. Ornelas, E. Trifoni, V. Antonucci, A.S. Arico, Int. J. Hydrogen Energy, 35, 5558 (2010).
 F. Jomard, J. Feraud, J.P. Caire, Int. J. Hydrogen Energy, 33, 1142 (2008).
 N.V. Dale, M.D. Mann, H. Salehfar. J. Power Sources, 185, 1348 (2008).
 Demin, E. Gorbova, P. Tsiakaras. J. Power Sources, 171, 205 (2007).
 H. Gorun, Int. J. Hydrogen Energy, 31, 29 (2006).
 M. Santarelli, P. Medina, M. Cali, Int. J. Hydrogen Energy, 34, 2519 (2009).
 S. Becker, V. Karri. doi:10.1016/j.ijhydene.2009.11.060.
 A.U. Chávez-Ramírez, R. Muñoz-Guerrero, S.M. Durón-Torres, M. Ferraro, G. Brunaccini, F. Sergi, V. Antonucci, L.G. Arriaga, 35, 12125 (2010).
 D. Graupe, “Principles of artificial neural networks”, ed. World Scientific, University of Illinois, Chicago USA (2007) , ISBN 10 981-270-624-0.
 J.C. Cruz, V. Baglio, J. Nanopart Res. DOI 10.1007/s11051-010-9917-2.
 M.Y. El-Sharkh, A. Rahman, M.S. Alam, J. Power Sources, 135, 88 (2004).
 Xiao-Juan Wu, Xin–Jian Zhu, Guang-Yi Chao, Heng-Yong Tu, J. Power Sources, 167, 145 (2007).
 A. Saengrung, A. Abtahi, A. Zilouchian, J. Power Sources, 172, 749 (2007).
 O. Poleyova, “Accelerated testing methodology of fuel cell stacks; 2nd Annual International symposium Fuel Cells Durability and performance 2006”. Miami Florida USA.