Abstract
A neural network with a feed forward topology and Bayesian regularization training algorithm is used to predict the austenite formation temperatures (Ac1 and Ac3) by considering the percentage of aloying elements in chemical composition of steel. The data base used here involves a large variety of diferent steel types such as structural steels, stainless steels, rail steels, spring steels, high temperature creep resisting steels and tool steels. Scater diagrams and mean relative error (MRE) statistical criteria are used to compare the performance of developed neural network with the results of Andrew s empirical equations and a feed forward neural network with “gradient descent with momentum’ training algorithm. The results showed that Bayesian regularization neural network has the best performance. Also, due to the satisfactory results of the developed neural network, it was used to investigate the effect of the chemical composition on Ac1 and Ac3 temperatures. Results are in accordance with materials science theories.
Similar content being viewed by others
References
K. W. Andrew, JISI 203 (1965) 721–727
P. Wang, Z. Y. Huang, M. Y. Zhang, X. W. Zhao, J. Iron Steel Res. Int. 15 (2008) No. 3, 87–91.
W. You, Y. X. Liu, B. Z. Bai, H. S. Fang. J. Iron Steel Res. Int. 15 (2008) No. 2, 87–90.
X. H. Gao, K. M. Qi, T. Y. Deng, C. L. Qiu, P. Zhou, X. B. Du, J. Iron Steel Res. Int. 13 (2006) No. 6, 71–73.
C. L. Tang, S. G. Wang, Q. H. Liang, W. Xu, J. Iron Steel Res. Int. 16 (2009) No. 5, 50–55.
L. A. Dobrzański, J. Trzaska, J. Mater. Process. Technol. 157–158 (2004) 107–113.
L. A. Dobrzański, J. Trzaska, J. Mater. Process. Technol. 155–156 (2004) 1950–1955.
L. Gavard, H. K. D. H. Bhadeshia, D. J. C. Mackay, S. Suzuki, J. Mater. Process. Technol. 12 (1996) 453–463.
W. You, W. Xu, B. Bai, H. Fang, Mater. Sci. Eng. A 419 (2008) 276–282.
MATLAB® Software, R2008a, Neural Network Toolbox, User’s Guide, 2008.
D. J. C. Mackay, Bayesian Methods for Adaptive Models, California Institute of Technology, California, 1991.
F. D. Foresee, M. T. Hagan, Proc. of the 1997 Int. Joint Conf. on Neural Networks 3 (1997) 1930–1935.
K. Hirschen, M. Schäfer, Comput. Methods Appl. Mech. Engrg. 195 (2006) 481–500.
M. Botlani Esfahani, M. R. Toroghinejad, A. R. Key Yeganeh, Mater. Des. 30 (2009) 3653–365.
L. Xu, J. Xing, S. Wei, Y. Zhang, R. Long, Mater. Sci. Eng. A 433 (2006) 251–256.
Atlas of Continuous Cooling Transformation Diagrams for Vanadium Steels, VanadiumInternational Technical Commitee, England, June, 1985.
M. T. Hagan, H. B. Demath, M. Beale, Neural Network Design, PWS Publications, Boston, 1996.
M. Rakhshkhorshid, S. H. Hashemi, IJMMNO. 4 (2013) No. 3, 238–251.
V. B. Ginzburg, R. Balas, Flat Rolling Fundamentals (Manufacturing Engineering and Materials Processing), Marcel Dekker Inc., New York, 2000.
W. D. Calister Jr., D. G. Rethwisch, Materials Science and Engineering: an Introduction, Eighth Edition, Wiley and Sons Inc., New Jersey, 2010.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Rakhshkhorshid, M., Teimouri Sendesi, SA. Bayesian Regularization Neural Networks for Prediction of Austenite Formation Temperatures (Ac1 and Ac3). J. Iron Steel Res. Int. 21, 246–251 (2014). https://doi.org/10.1016/S1006-706X(14)60038-8
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1016/S1006-706X(14)60038-8