期刊
KSCE JOURNAL OF CIVIL ENGINEERING
卷 19, 期 3, 页码 611-620出版社
KOREAN SOCIETY OF CIVIL ENGINEERS-KSCE
DOI: 10.1007/s12205-013-0315-z
关键词
driven pile; Flap number; bearing capacity; neural network; genetic programming
A new method was proposed to predict the compressive bearing capacity of driven piles based on the number of hammer strikes in the last one meter of pile penetration (known here as Flap number). To collect the data, a literature review was done on technical publications and pile driving record reports that were accessible to the authors at the time of publication. The data of a hundred driven piles including Flap number, basic properties of the surrounding soil, pile geometry, and pile-soil friction angle was collected. These data were initially used in the artificial neural network to establish a relation for predicting pile capacity. Subsequently, by using genetic programing and linear regression, equations for determining pile bearing capacity with respect to the Flap number, soil parameters, and pile geometries were proposed. Finally, the performance of all applied methods in predicting the pile bearing capacity were compared. The utmost importance was given to the comparison of the accuracy of the three models as well as the error estimation.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据