A comparative study of some network approaches to predict the effect of the reinforcement content on the hot strength of Al–base composites

Abstract Due to the nonlinear complex effect of the reinforcement content and the deformation conditions such as temperature and strain rate on the flow stress, the existing models especially those dependent on the activation energy are not suitable to predict the hot deformation behavior. To achieve this purpose, it was decided to use the property of some of the existing models such as radial-base function network (RBF), multi-layer perceptron (MLP) network, and neuro-fuzzy network to predict the nonlinear behavior in the stress–strain behavior of the material. The results showed that the neuro-fuzzy network is the best tool to predict the hot deformation behavior of Al–base composites with different reinforcement content (5, 10, 15, and 20%) of Al 2 O 3 particles that have an average particle size of 25 μm at different deformation conditions since the reinforcement content and the deformation conditions have a nonlinear complex effect on the flow stress.

[1]  John G. Lenard,et al.  Using neural networks to predict parameters in the hot working of aluminum alloys , 1999 .

[2]  L. C. Davis.  Third-order bounds on the elastic moduli of metal-matrix composites , 1991 .

[3]  Z. Hashin.  Analysis of Composite Materials—A Survey , 1983 .

[4]  Jyh-Shing Roger Jang,et al.  ANFIS: adaptive-network-based fuzzy inference system , 1993, IEEE Trans. Syst. Man Cybern..

[5]  D. Broomhead,et al.  Radial Basis Functions, Multi-Variable Functional Interpolation and Adaptive Networks , 1988 .

[6]  Graeme W. Milton,et al.  Bounds on the Electromagnetic, Elastic, and Other Properties of Two-Component Composites , 1981 .

[7]  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 .

[8]  Salvatore Torquato,et al.  Random Heterogeneous Media: Microstructure and Improved Bounds on Effective Properties , 1991 .

[9]  Lingxue Kong,et al.  Modelling the Effect of Carbon Content on Hot Strength of Steels Using a Modified Artificial Neural Network , 1998 .

[10]  Snehasis Mukhopadhyay,et al.  Selecting an artificial neural network for efficient modeling and accurate simulation of the milling process , 2002 .

[11]  L. C. Davis,et al.  Residual stresses and their effects on deformation , 1993, Metallurgical and Materials Transactions A.

[12]  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 .

[13]  Sheng Chen,et al.  Practical identification of NARMAX models using radial basis functions , 1990 .

[14]  John G. Lenard,et al.  A comparative study of artificial neural networks for the prediction of constitutive behaviour of HSLA and carbon steels , 1996 .

[15]  C. Micchelli.  Interpolation of scattered data: Distance matrices and conditionally positive definite functions , 1986 .

[16]  Saman K. Halgamuge,et al.  Neural networks in designing fuzzy systems for real world applications , 1994 .

[17]  J. Whittenberger,et al.  High-temperature slow-strain-rate compression studies on CoAI-TiB2 composites , 1990 .