Modeling the Mechanical Behavior of Carbonate Sands Using Artificial Neural Networks and Support Vector Machines
AbstractCarbonate sands that are specific soils have some unusual characteristics, such as particle crushability and compressibility, that distinguish their behavior from other types of soil. Because of their large diversity, they have a wide range of mechanical behavior. Recently, there have been m...
Gespeichert in:
Veröffentlicht in: | International journal of geomechanics 2016-02, Vol.16 (1) |
---|---|
Hauptverfasser: | , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | AbstractCarbonate sands that are specific soils have some unusual characteristics, such as particle crushability and compressibility, that distinguish their behavior from other types of soil. Because of their large diversity, they have a wide range of mechanical behavior. Recently, there have been many attempts to predict the mechanical behavior of carbonate sands, but all these attempts have been focused on experimental and case studies of some specific soils, and there is still no unique method that can consider all types of carbonate sands behavior and describe their various aspects. In the present study, two artificial intelligence-based models, namely artificial neural networks and support vector machines are used together and comparatively to predict the mechanical behavior of different carbonate sands. The models were trained and tested using a database that included results from a comprehensive set of triaxial tests on three carbonate sands. The predictions of the proposed models were compared with the experimental results. The comparison of the results indicates that the proposed approaches were accurate and reliable in representing the mechanical behavior of various carbonate sands. |
---|---|
ISSN: | 1532-3641 1943-5622 |
DOI: | 10.1061/(ASCE)GM.1943-5622.0000509 |