Optimization of Three-dimensional Electrical Performance in a Solid Oxide Fuel Cell Stack by a Neural Network
Commenced in January 2007
Frequency: Monthly
Edition: International
Paper Count: 32769
Optimization of Three-dimensional Electrical Performance in a Solid Oxide Fuel Cell Stack by a Neural Network

Authors: Shih-Bin Wang, Ping Yuan, Syu-Fang Liu, Ming-Jun Kuo

Abstract:

By the application of an improved back-propagation neural network (BPNN), a model of current densities for a solid oxide fuel cell (SOFC) with 10 layers is established in this study. To build the learning data of BPNN, Taguchi orthogonal array is applied to arrange the conditions of operating parameters, which totally 7 factors act as the inputs of BPNN. Also, the average current densities achieved by numerical method acts as the outputs of BPNN. Comparing with the direct solution, the learning errors for all learning data are smaller than 0.117%, and the predicting errors for 27 forecasting cases are less than 0.231%. The results show that the presented model effectively builds a mathematical algorithm to predict performance of a SOFC stack immediately in real time. Also, the calculating algorithms are applied to proceed with the optimization of the average current density for a SOFC stack. The operating performance window of a SOFC stack is found to be between 41137.11 and 53907.89. Furthermore, an inverse predicting model of operating parameters of a SOFC stack is developed here by the calculating algorithms of the improved BPNN, which is proved to effectively predict operating parameters to achieve a desired performance output of a SOFC stack.

Keywords: a SOFC stack, BPNN, inverse predicting model of operating parameters, optimization of the average current density

Digital Object Identifier (DOI): doi.org/10.5281/zenodo.1074291

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1312

References:


[1] Yakabe, H., Hishinuma, M., Uratani, M., Matsuzaki, Y., Yasuda, I., "Evaluation and modeling of performance of anode-supported solid oxide fuel cell", J. Power Sources, vol. 86, 2000, pp. 423-431.
[2] Yakabe, H., Ogiwara, T., Hishinuma, M., Yasuda, I., "3-D model calculation for planar SOFC", J. Power Sources, vol. 102, 2001, pp. 144-154.
[3] Recknagle, K.P., Williford, R.E., Chick, L.A., Rector, D.R., Khaleel, M.A., "Three-dimensional thermo-fluid electrochemical modeling of plannar SOFC stacks", J. Power Sources, vol. 113, 2003, pp. 109-114.
[4] Beale, S.B., Lin, Y., Zhubrin, S.V., Dong, W., "Computer methods for performance prediction in fuel cells", J. Power Sources, vol. 118, 2003, pp. 79-85.
[5] Iwata, M., Hikosaka, T., Morita, M., Iwanari, T., Ito, K., Onda, K., Esaki, Y., Sakaki, Y., Nagata, S., "Performance analysis of planar-type unit SOFC considering current and temperature distributions, Solid State Ionics", vol. 132, 2000, pp. 297-308.
[6] J.J. Huang, C.K. Chen and D.Y. Lai, J. Power Sources 140 (2005), pp. 235-242.
[7] V.M. Janardhanan, V. Heuveline and O. Deutschmann, J. Power Sources 172 (2007), pp. 296-307.
[8] T. Araki, T. Ohba, S. Takezawa, K. Onda and Y. Sakaki, J. Power Sources 158 (2006), pp. 52-59.
[9] H. Hirata and M. Hori, J. Power Sources 63 (1996), pp. 115-120.
[10] S.F. Liu, H.S. Chu and P. Yuan, J. Power Sources 161 (2006), pp. 1030-1040.
[11] P. Yuan and S.F. Liu, Numer. Heat Transf. A: Appl. 51 (2007), pp. 941-957
[12] P. Costamagna, E. Arato, E. Achenbach and U. Reus, J. Power Sources 52 (1994), pp. 243-249.
[13] R.J. Boersma and N.M. Sammes, J. Power Sources 66 (1997), pp. 41-45.
[14] T. Okada, S. Matsumoto, M. Matsumura, M. Miyazaki and M. Umeda, J. Power Sources 162 (2006), pp. 1029-1035.
[15] Ping Yuan, J. Power Sources 185 (2008), pp. 381-391.
[16] Arriagada, J., Olausson, P., Selimovic, A., "Artificial neural network simulator for SOFC performance prediction", J. Power Sources, vol. 112, 2002, pp. 54-60.
[17] J.H. Koh, H.K. Seo, Y.S. Yoo and H.C. Lim, Chem. Eng. J. 87 (2002), pp. 367-379.
[18] L.J.M.J. Blomen and M.N. Mugerwa, Fuel Cell Systems, Plenum Press, New York (1993) pp. 73-75.
[19] S.H. Chan, K.A. Khor and Z.T. Xia, J. Power Sources 93 (2001), pp. 130-140.
[20] J. Larminie and A. Dicks, Fuel Cell Systems Explained (1st ed.), Wiley, West Sussex (2000) p. 53.
[21] R. Maric, S. Ohara, T. Fukui, H. Yoshida, M. Nishimura, T. Inagaki and K. Miura, J. Electrochem. Soc. 146 (1999), pp. 2006-2010.
[22] A.L. Hines and R.N. Maddox, Mass Transfer Fundamentals and Applications, Prentice-Hall, New Jersey (1985) pp. 17-59.
[23] Wang, S.B., Wu, C.F., "Selections of working conditions for creep feed grinding. Part (III): avoidance of the work piece burning by using improved BP neural network", Int. J. Adv. Manuf. Technol., vol. 28, 2006, pp. 31-37.
[24] Rangwala, S., Dornfeld, D., "Learning and optimization of machining operations using computing abilities of neural networks", IEEE Trans. on Systems, man and Cybernetics-, vol.19, No. 2, 1989, pp. 299-314.
[25] Peace, G., Taguchi Method: A Hands-on Approach, Addison-Wesley, Reading, MA,1993.