Prediction of the Flow Stress of 00Cr17Ni14Mo2 Steel during Hot Deformation Using a BP Artificial Neural Network

Abstract:

The knowledge of the flow behavior of metals during hot deformation is of great importance in determining the optimum forming conditions. In this paper, the flow stress of 00Cr17Ni14Mo2 (ANSI 316L) austenitic stainless steel in elevated temperature is measured with compression deformation tests. The temperatures at which the steel is compressed are 800-1100°C with strain rates of 0.01-1s-1. With the experiment result as the training set, the flow stress of 00Cr17Ni14Mo2 steel during hot deformation is predicted using a BP artificial neural network. The architecture of the network includes three input parameters, one output parameter and two hidden layers with 5 neurons in first layer and 6 neurons in second layer. Compared with the regression method, the prediction using the BP artificial neural network has higher efficiency and accuracy.

• Periodical: Advanced Materials Research (Volumes 181-182)
• Pages: 979-982
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.181-182.979
• Online since: January 2011
• Authors: Nan Hai Hao, Jiu Shi Li
• Keywords: Artificial Neural Network (ANN), Flow Stress, Stainless Steel (SS)

References

[1] N. Tounsi, J. Vincenti, A. Otho and M.A. Elbestawi: Int. J. Mach. Tools & Manu., Vol. 42 (2002), p.1373.

[2] G. Angella, B. P. Wynne, W. M. Rainforth and J. H. Beynon: Mater. Sci. Eng. A, Vol. 475 (2008), p.257.

[3] M. P. Phaniraj and A. K. Lahiri: J. Mater. Process. Technol., Vol. 141 (2003), p.219.

[4] R. H. Wu, J. T. Liu, H. B. Chang, T. Y. Hsu and X. Y. Ruan: J. Mater. Process. Technol., Vol. 116 (2001), p.211.

[5] S. Haykin: Neural Networks, A comprehensive Foundation (Prentice-Hall, USA 1994).