Simulating the stress–strain behavior of Georgia kaolin via recurrent neuronet approach

Abstract Many geotechnical engineering problems, such as slope stability analysis, require understanding of the constitutive behavior of clay under plane strain loading condition. Although studies on the mechanism of clay behavior had been carried out for many years, current theoretical accomplishments still cannot provide reliable simulations for clay under plane strain loading encountered in engineering practice. Traditional constitutive modeling, which involves building mathematical formulations with parameters determined from clay tests, seems to be an attractive approach. However, the complexity of the formulation, the large number of parameters involved, and the difficulty and time required to conduct clay tests make this approach impractical. As a result, there are no available reliable simulation models that can efficiently characterize the clay response under plane strain loading. Lack of appropriate simulation (constitutive) model for clay under plane strain loading greatly hinders the analysis, design, and construction with this material in engineering practice. Therefore, it is worthy to find another way to study the clay response under plane strain loading. In this paper, a recurrent neuronet-based model was developed in order to simulate clay behavior under plane strain loading conditions. This model was then used to investigate the effects of loading rate and stress history on clay response. The predicted responses showed that the developed neuronet-based model was successfully able to qualitatively assess the impact of strain rate and stress history on clay behavior. Accordingly, it is found that clay behavior is more sensitive to its consolidation stress history than to its strain loading rate.

[1]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[2]  Les E. Atlas,et al.  Recurrent neural networks and robust time series prediction , 1994, IEEE Trans. Neural Networks.

[3]  Gioacchino Viggiani,et al.  Strain Localization and Undrained Steady State of Sand , 1996 .

[4]  A. Casacrande,et al.  Effect of Rate of Loading on the Strength of Clays and Shales at Constant Water Content , 1951 .

[5]  James L. McClelland.  Explorations In Parallel Distributed Processing , 1988 .

[6]  C. L. Giles,et al.  Dynamic recurrent neural networks: Theory and applications , 1994, IEEE Trans. Neural Networks Learn. Syst..

[7]  Yacoub M. Najjar,et al.  Characterizing the Fatigue Life of Asphalt Concrete , 2006 .

[8]  Stein Sture,et al.  Sand Shear Band Thickness Measurements by Digital Imaging Techniques , 1999 .

[9]  Gioacchino Viggiani,et al.  Undrained Shear Band Deformation in Granular Media , 1997 .

[10]  James H. Garrett,et al.  Knowledge-Based Modeling of Material Behavior with Neural Networks , 1992 .

[11]  Laurene V. Fausett,et al.  Fundamentals Of Neural Networks , 1994 .

[12]  Ken-ichi Funahashi,et al.  On the approximate realization of continuous mappings by neural networks , 1989, Neural Networks.

[13]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[14]  Amir F. Atiya,et al.  Application of the recurrent multilayer perceptron in modeling complex process dynamics , 1994, IEEE Trans. Neural Networks.

[15]  Dayakar Penumadu,et al.  Triaxial compression behavior of sand and gravel using artificial neural networks (ANN) , 1999 .

[16]  Yacoub Najjar.  Characterizing the 3D Stress-Strain Behavior of Sandy Soils: A Neuro-Mechanistic Approach , 2000 .

[17]  Xiping Wu.  Neural network-based material modeling , 1992 .

[18]  Jeffrey L. Elman,et al.  Finding Structure in Time , 1990, Cogn. Sci..

[19]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[20]  D. E. Rumelhart,et al.  Learning internal representations by back-propagating errors , 1986 .