Predicting Peak Stress and Strain of Concrete under Triaxial Stress using Neural Network Models
| Autor: | Ren, Wei Chang, 任偉誠 |
|---|---|
| Rok vydání: | 2012 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 100 It is well known that concrete is a kind of heterogeneity and anisotropic nonlinear materials. The nonlinear behavior of concrete under triaxial stress is very complicated; modeling its behavior is a hard task. Recently, it has shown that artificial neural network-based modeling is an alternative method for modeling complex nonlinear relationship. In this study, therefore, the standard back-propagation neural network (BPN) was used for modeling and estimating the peak stress and strain of plain concrete under triaxial stress. The results show that the BPN models give reasonable predictions of the peak stress and strain of plain concrete under triaxial stress. In addition, the BPN models provide better accuracy than the existing parametric models. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|