An artificial neural network approach to multiphase continua constitutive modeling

Constitutive equations describe intrinsic relationships among sets of material system parameters. This study utilizes artificial neural networks in place of a traditional micromechanical approach to calculate the global (macroscopic) elastic properties of composite materials given the local (microscopic) properties and local geometry. This approach is shown to be more computationally efficient than conventional numerical micromechanical approaches. An eight sub-celled representative volume element is used for the local geometry. Multi target artificial neural networks (MTANNs) and single target artificial neural networks are studied for applicability in predicting the global properties. The best performing MTANN achieves a precision of 9%. The single target artificial neural networks (STANNs) perform best and predicts the global properties within a target error of 5.3%. The computation time is 1.8 s for all six STANNs to predict six global properties for 19,683 different microstructures.

[1]  T. Ikeda.  Fundamentals of piezoelectricity , 1990 .

[2]  Jacob Aboudi,et al.  Micromechanical Prediction of the Effective Coefficients of Thermo-Piezoelectric Multiphase Composites , 1998 .

[3]  L. E. Malvern.  Introduction to the mechanics of a continuous medium , 1969 .

[4]  Modeling of the effect of the hot-deformation parameters on the strength of Al-base metal–matrix composites by the use of a radial-base function (RBF) network , 2001 .

[5]  J. T. Liu,et al.  PREDICTION OF FLOW STRESS OF HIGH-SPEED STEEL DURING HOT DEFORMATION BY USING BP ARTIFICIAL NEURAL NETWORK , 2000 .

[6]  Martin T. Hagan,et al.  Neural network design , 1995 .

[7]  George Z. Voyiadjis,et al.  SIMULATED MICROMECHANICAL MODELS USING ARTIFICIAL NEURAL NETWORKS , 2001 .

[8]  I A Basheer,et al.  Artificial neural networks: fundamentals, computing, design, and application. , 2000, Journal of microbiological methods.

[9]  Ji Zhong,et al.  Acquiring the constitutive relationship for a thermal viscoplastic material using an artificial neural network , 1996 .

[10]  David J. C. MacKay,et al.  A Practical Bayesian Framework for Backpropagation Networks , 1992, Neural Computation.

[11]  Anthony G. Evans,et al.  Methodology for Relating the Tensile Constitutive Behavior of Ceramic‐Matrix Composites to Constituent Properties , 1994 .

[12]  A New Inverse Method of Elastic Constants for a Fibre-Reinforced Composite Plate from Laser-Based Ultrasonic Lamb Waves , 2001 .

[13]  Martin T. Hagan,et al.  Gauss-Newton approximation to Bayesian learning , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[14]  Frank W. Zok,et al.  The physics and mechanics of fibre-reinforced brittle matrix composites , 1994, Journal of Materials Science.

[15]  I. Jalham.  Modeling capability of the artificial neural network (ANN) to predict the effect of the hot deformation parameters on the strength of Al-base metal matrix composites , 2003 .

[16]  Abhijit Mukherjee,et al.  Artificial neural networks for predicting the macromechanical behaviour of ceramic-matrix composites , 1996 .

[17]  Wray L. Buntine,et al.  Bayesian Back-Propagation , 1991, Complex Syst..

[18]  David J. C. MacKay,et al.  Bayesian Model Comparison and Backprop Nets , 1991, NIPS.

[19]  David J. C. MacKay,et al.  Bayesian Interpolation , 1992, Neural Computation.