全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

Development of a Neural Network Simulator for Studying the Constitutive Behavior of Structural Composite Materials

DOI: 10.1155/2013/147086

Full-Text   Cite this paper   Add to My Lib

Abstract:

This paper introduces a recent development and application of a noncommercial artificial neural network (ANN) simulator with graphical user interface (GUI) to assist in rapid data modeling and analysis in the engineering diffraction field. The real-time network training/simulation monitoring tool has been customized for the study of constitutive behavior of engineering materials, and it has improved data mining and forecasting capabilities of neural networks. This software has been used to train and simulate the finite element modeling (FEM) data for a fiber composite system, both forward and inverse. The forward neural network simulation precisely reduplicates FEM results several orders of magnitude faster than the slow original FEM. The inverse simulation is more challenging; yet, material parameters can be meaningfully determined with the aid of parameter sensitivity information. The simulator GUI also reveals that output node size for materials parameter and input normalization method for strain data are critical train conditions in inverse network. The successful use of ANN modeling and simulator GUI has been validated through engineering neutron diffraction experimental data by determining constitutive laws of the real fiber composite materials via a mathematically rigorous and physically meaningful parameter search process, once the networks are successfully trained from the FEM database. 1. Introduction Constitutive law (the stress/strain relationship) of the materials determines the mechanical response under quasistatic loading in both elastic and plastic regions and is the most fundamental information for structural applications [1]. While the elastic behavior can be well characterized based on the materials’ intrinsic physical properties ( : Young’s modulus, : Shear modulus, : Poisson’s ratio, etc. ( (or ) and are enough for isotropic materials. However, single crystal stiffness (or compliance) tensor needs to be applied for non-isotropic materials, depending on the crystal structure)), the plastic flow is more complicated and depends on the extrinsic material properties (dislocation density, grain size and orientation, available slip systems, etc.) and experimental conditions (temperature, strain rate, sample size and geometry, etc.). Furthermore, since most load bearing materials are not monolithic forms of a single element, obtaining the correct constitutive laws of the composite systems is not a trivial task. The engineering neutron diffractometer (e.g., SMARTS [2] in Los Alamos National Laboratory (LANL) or VULCAN [3] in Oak Ridge

References

[1]  R. Hill, “The essential structure of constitutive laws for metal composites and polycrystals,” Journal of the Mechanics and Physics of Solids, vol. 15, no. 2, pp. 79–95, 1967.
[2]  M. A. M. Bourke, D. C. Dunand, and E. Ustundag, “SMARTS—a spectrometer for strain measurement in engineering materials,” Applied Physics A, vol. 74, pp. S1707–S1709, 2002.
[3]  X. L. Wang, T. M. Holden, G. Q. Rennich et al., “VULCAN—the engineering diffractometer at the SNS,” Physica B, vol. 385-386, pp. 673–675, 2006.
[4]  C. N. Tomé, “Self-consistent polycrystal models: a directional compliance criterion to describe grain interactions,” Modelling and Simulation in Materials Science and Engineering, vol. 7, no. 5, pp. 723–738, 1999.
[5]  J. J. Hopfield, “Neural networks and physical systems with emergent collective computational abilities,” Proceedings of the National Academy of Sciences of the United States of America, vol. 79, no. 8, pp. 2554–2558, 1982.
[6]  D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning representations by back-propagating errors,” Nature, vol. 323, no. 6088, pp. 533–536, 1986.
[7]  I. Aleksander and H. Morton, An Introduction to Neural Computing, Van Nostrand Reinhold Co., New York, NY, USA, 1990.
[8]  C. M. Bishop, Neural Networks for Pattern Recognition, Oxford University Press, Oxford, UK, 1995.
[9]  K. Swingler, Applying Neural Networks: A Practical Guide, Academic Press, 1996.
[10]  H. Ceylan, Analysis and Design of Concrete Pavement Systems Using Artificial Neural Networks, University of Illinois at Urbana-Champaign, Urbana, Ill, USA, 2002.
[11]  B. Clausen, S. Y. Lee, E. üstündag, C. C. Aydiner, R. D. Conner, and M. A. M. Bourke, “Compressive yielding of tungsten fiber reinforced bulk metallic glass composites,” Scripta Materialia, vol. 49, no. 2, pp. 123–128, 2003.
[12]  B. Denizer, E. Ustundag, H. Ceylan, L. Li, and S. Y. Lee, “Engineering neutron diffraction data analysis with inverse neural network modeling,” Materials Science Forum. In press.
[13]  S. L. Ross, “Parametric investigation of artificial neural network development to study composites,” unpublished work.
[14]  B. Denizer, Artificial Neural Network Analysis of the Mechanical Properties of Tungsten Fiber/Bulk Metallic Glass Matrix Composites via Neutron Diffraction and Finite Element Modeling, Iowa State University, Ames, Iowa, USA, 2008.
[15]  D. Dragoi, E. üstündag, B. Clausen, and M. A. M. Bourke, “Investigation of thermal residual stresses in tungsten-fiber/bulk metallic glass matrix composites,” Scripta Materialia, vol. 45, no. 2, pp. 245–252, 2001.

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133