Abstract
In this study, a Back-propagation neural network (BPN) is employed to conduct an approximation of a true stress-strain curve using the load-displacement experimental data of DP590, a high-strength material used in automobile bodies and chassis. The optimized interconnection weights are obtained with hidden layers and output layers of the BPN through intelligent learning and training of the experimental data; by using these weights, a mathematical model of the material’s behavior is suggested through this feed-forward neural network. Generally, the material properties from the tensile test cannot be acquired until the fracture regions, since it is difficult to measure the cross-section area of a specimen after diffusion necking. For this reason, the plastic properties of the true stress-strain are extrapolated using the weighted-average method after diffusion necking. The accuracies of BPN-based meta-models for predicting material properties are validated in terms of the Root mean square error (RMSE). By applying the approximate material properties, the reliable finite element solution can be obtained to realize the different shapes of the finite element models. Furthermore, the sensitivity analysis of the approximate meta-model is performed using the first-order approximate derivatives of the BPN and is compared with the results of the finite difference method. In addition, we predict the tension velocity’s effect on the material property through a first-order sensitivity analysis.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
P. W. Bridgman, Studies in large plastic flow and fracture, McGraw Hill Book Company Inc., New York (1952).
Y. Ling, Uniaxial true stress-strain after necking, AMP Journal of Technology, 5 (1996) 37–48.
H. C. Hyun, J. H. Lee and H. Lee, Mathematical expressions for stress-strain curve of metallic material, Korean Society of Mechanical Engineers, 32 (2008) 21–28.
H. Sudarsana Rao, V. G. Ghorpade and A. Mukherjee, A genetic algorithm based back propagation network for simulation of stress–strain response of ceramic-matrixcomposites, Computer and Structure, 84 (2006) 330–339.
S. U. Lee, Prediction of springback of DP590 sheet metal using Yoshida-Uemori model, M.S. Thesis, Department of Mechanical Engineering, Kyungpook National University, Korea (2012).
M. El-Zeghayar, T. Topper, J. Bonnen and R. Sohmshetty, Effective strain-fatigue life of dual phase 590 steel, Proceedings of the 12th International Conference on Fracture, Ottawa, Canada (2009) 1–10.
J. Lee and P. Hajela, Parallel genetic algorithm implementation in multidisciplinary rotor blade design, Journal of Aircraft, 33 (5) (1996) 962–969.
H. Kim and J. Lee, Parameter analysis and optimization of paper feeding devices to minimize jamming and simultaneous feeding of multiple pages, Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 225 (11) (2011) 2673–2684.
H. Jeong, J. H. Yun, J. C. Ko, I. H. Lee and J. Lee, Approximate analysis of first order sensitivity using backpropagation neural networks, Proceedings of the Fall Annual Meeting of Korean Society of Mechanical Engineers, Changwon, Korea, November (2012).
KS-B0801-13B, Korean Standards Information Center, Korean Industrial Standard, Seoul, Korea (2007).
INSTRON 5569, Instron Corporation, Norwood, MA (2006).
KS B 0802, Korean Standards Information Center, Korean Industrial Standard, Seoul, Korea (2003).
S. P. Keeler and W. A. Back, Plastic instability and fracture in sheet stretched over rigid punches, ASM Transactions Quarterly, 56 (1) (1963) 25–48.
Y. Seo, H. C. Hyun, H. Lee and N. Kim, Forming limit diagrams of Zircaloy-4 and Zirlo sheets for stamping of spacer grids of nuclear fuel rods, Korean Society of Mechanical Engineers, 35 (8) (2011) 889–897.
HyperMesh User Guide, HyperMesh Version 11.0, Altair Engineering Inc., Troy, MI (2012).
ABAQUS User’s Manual, Version 6.12, Dassault Systemes Simulia, Inc., Providence, RI, USA (2012).
H. C. Hyun, M. Kim, S. Bang and H. Lee, On acquiring true stress-strain curves for sheet specimen using tensile test and fe analysis based on a local necking criterion, Journal of Materials Research, 29 (2014) 695–707.
J. Lee, H. Jeong, D. H. Choi, V. Volovoi and D. Mavris, An enhancement of constraint feasibility in BPN based approximate optimization, Computer Method in Applied Mechanics and Engineering, 196 (2007) 2147–2160.
R. J. Hyndman and A. B. Koehler, Another look at measures of forecast accuracy, International Journal of Forecasting, 22 (4) (2006) 679–688.
P. Han and J. Lee, A response surface based sequential approximate optimization using constraint-shifting analogy, Journal of Mechanical Science and Technology, 23 (2009) 2903–2912.
Author information
Authors and Affiliations
Corresponding author
Additional information
Jaehyeok Doh is a Ph.D. student in Mechanical Engineering at Yonsei University. His research interests are on the field of structural analysis, finite element method and probabilistic design optimization.
Jongsoo Lee received B.S. in Mechanical Engineering at Yonsei University, Korea in 1988 and Ph.D. in Mechanical Engineering at Rensselaer Polytechnic Institute, Troy, NY in 1996. After a research associate at Rensselaer Rotorcraft Technology Center, he is a professor of Mechanical Engineering at Yonsei University. His research interests include multi- disciplinary/ multi-physics/multi-scale design optimization and reliability- based robust engineering design with applications to structures, structural dynamics, fluid-structure interactions and flow induced noise and vibration problems.
Rights and permissions
About this article
Cite this article
Doh, J., Lee, S.U. & Lee, J. Back-propagation neural network-based approximate analysis of true stress-strain behaviors of high-strength metallic material. J Mech Sci Technol 30, 1233–1241 (2016). https://doi.org/10.1007/s12206-016-0227-1
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12206-016-0227-1