سال انتشار: ۱۳۸۶

محل انتشار: پنجمین کنفرانس بین المللی زلزله شناسی و مهندسی زلزله

تعداد صفحات: ۹

نویسنده(ها):

Ghateh – (M. Sc). International Institute of Earthquake Engineering and Seismology
Shafiee – (Ph. D). International Institute of Earthquake Engineering and Seismology

چکیده:

Compacted aggregate-clay mixtures are successfully used as the core of embankment dams. These materials are usually broadly graded and encompass clay as the main body, and sand, gravel, cobble and even boulder which are floating in the clay matrix. The objective of this paper is to develop a mathematical model for predicting the cyclic deformation properties of aggregate-clay mixtures tested previously by Jafari and Shafiee (2004). A mathematical modeling, which has been shown to have some degree of success, is based on the data alone to determine the structure and parameters of the model. The technique is known as artificial neural networks (ANNs) and is well suited to model complex problems where the relationship between the model variables is unknown. A neural network is a massively parallel distributed processor made up of simple processing units, which has a natural propensity for storing experiential knowledge and making it available for use.
Over the last few years, the use of artificial neural networks (ANNs) has increased in many areas of engineering. In particular, ANNs have been applied to many geotechnical engineering problems and have demonstrated some degree of success. A review of the literature reveals that ANNs has been used successfully in pile capacity prediction, modeling soil behavior, liquefaction, etc. In this paper two MLP models with different architectures are utilized for predicting damping ratio, shear modulus and pore pressure of aggregate-clay mixtures. The reliability of the models is tested using cross validation technique. The data used for training the networks is based on the laboratory tests for determining the dynamic properties of aggregate-clay mixtures. Finally the importance of each input parameter is determined using Garson’s approach.