Abstract
The prediction of the mechanical properties of hot-rolled strips is a very complex, highly dimensional and nonlinear problem, and the published models might lack reliability, practicability and generalization. Thus, a new model was proposed for predicting the mechanical properties of hot-rolled strips by deep learning. First, the one-dimensional numerical data were transformed into two-dimensional data for expressing the complex interaction between the influencing factors. Subsequently, a new convolutional network was proposed to establish the prediction model of tensile strength of hot-rolled strips, and an improved inception module was introduced into this network to abstract features from different scales. Many comparative experiments were carried out to find the optimal network structure and its hyperparameters. Finally, the prediction experiments were carried out on different models to evaluate the performance of the new convolutional network, which includes the stepwise regression, ridge regression, support vector machine, random forest, shallow neural network, Bayesian neural network, deep feed-forward network and improved LeNet-5 convolutional neural network. The results show that the proposed convolutional network has better prediction accuracy of the mechanical properties of hot-rolled strips compared with other models.
Similar content being viewed by others
References
S.M. Azimi, D. Britz, M. Engstler, M. Fritz, F. Mücklich, Sci. Rep. 8 (2018) 2128.
W.G. Li, W. Yang, Y.T. Zhao, G. Xu, X.H. Liu, J. Iron Steel Res. Int. 26 (2019) 230–241.
W.G. Li, W. Yang, Y.T. Zhao, H.F. Hu, J. Iron Steel Res. Int. 30 (2018) 302–308.
G. Khalaj, T. Azimzadegan, M. Khoeini, M. Etaat, Neural Comput. Appl. 23 (2013) 2301–2308.
X.Y. Sui, Z.M. Lv, J. Int. Adv. Manuf. 85 (2016) 1395–1403.
S.W. Wu, G.M. Cao, X.G. Zhou, N.A. Shi, Z.Y. Liu, ISIJ Int. 57 (2017) 1213–1220.
S.W. Wu, J.K. Ren, X.G. Zhou, G.M. Cao, Z.Y. Liu, J. Yang, Trans. Indian Inst. Met. 72 (2019) 1277-1288.
S.W. Wu, Z.Y. Liu, X.G. Zhou, N.A. Shi, J. Iron Steel Res. Int. 28 (2016) 1–4.
Y.H. Zhao, Y. Weng, N.Q. Peng, G.B. Tang, Z.D. Liu, J. Iron Steel Res. Int. 20 (2013) No. 7, 9–15.
A. Noroozi, M. Ayaz, N.B. Mostafa Arab, D. Mirahmadi Khaki, Metall. Res. Technol. 110 (2013) 359–371.
L. Čiripová, E. Hryha, E. Dudrová, A. Výrostková, Mater. Des. 35 (2012) 619–625.
D. Šimek, A. Oswald, R. Schmidtchen, M. Motylenko, G. Lehmann, D. Rafaja, Steel Res. Int. 85 (2014) 1369–1378.
A.A. dos Santos, R. Barbosa, Steel Res. Int. 81 (2010) 55–63.
C.Z. Zhang, B.M. Gong, C.Y. Deng, D.P. Wang, Mater. Sci. Eng. A 685 (2017) 310–316.
M. Alibeyki, H. Mirzadeh, M. Najafi, A. Kalhor, J. Mater. Eng. Perform. 26 (2017) 2683–2688.
W.Y. Xie, Y.S. Li, X.P. Jia, Neurocomputing 312 (2018) 372–381.
O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z.H. Huang, A. Karpathy, A. Khosla, M. Bernstein, A.C. Berg, F.F. Li, Int. J. Comput. Vision 115 (2015) 211-252.
A. Krizhevsky, I. Sutskever, G.E. Hinton, in: P. Bartlett (Eds.), NIPS. Conference and Workshop on Neural Information Processing Systems, Neural Information Processing Systems Foundation, Inc., Lake Tahoe, USA, 2012, pp. 1106–1114.
T. Hu, M. Yang, W.Q. Yang, A.S. Li, Int. J. Mach. Learn. Cyb. 10 (2019) 1909–1924.
M. Abadi, in: M. Abadi (Eds.), ICFP. The 21st ACM SIGPLAN International Conference on Functional Programming, Association for Computing Machinery, New York, USA, 2016.
W. Yang, W.G. Li, Y.T. Zhao, B. Yan, W. Wang, Iron and Steel 53 (2018) No. 3, 44–49.
N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, R. Salakhutdinov, J. Mach. Learn Res. 15 (2014) 1929–1958.
Acknowledgements
This research is supported by National Natural Science Foundation of China (51774219).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Li, Wg., Xie, L., Zhao, Yt. et al. Prediction model for mechanical properties of hot-rolled strips by deep learning. J. Iron Steel Res. Int. 27, 1045–1053 (2020). https://doi.org/10.1007/s42243-020-00450-9
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s42243-020-00450-9