基于人工神经网络的静力压桩沉桩阻力计算模拟  

Simulation for Penetrating Resistance of Jacket Pile Based on Artifical Neural Network

作　　者：张明义[1] 周常春[2] 邓安福[2] 

机构地区：[1]青岛建筑工程学院土木系,青岛266033 [2]重庆大学土木工程学院,重庆400045

出　　处：《工程勘察》2001年第6期8-11,共4页Geotechnical Investigation & Surveying

摘　　要：本文将人工神经网络用于计算静力压入桩的沉桩阻力 ,编制了以静力触探资料为主要输入因素的计算程序 ,程序可以对比显示计算和实测曲线。对不同地区的数根桩进行了训练和计算 ,计算结果令人满意。本文的工作证明人工神经网络可以模拟桩的贯入特性 ,所作的计算具有工程实用性。A computer program is compiled herein by taking the data of static sounding as the major input for calculating the penetrating resistance of jacked pile with the aid of artificial neural network.Both calculated curves and tested ones can be displayed by the program. Several piles of different areas trained and calculated with satisfactory results. It is verified that artificial neural network can simulate the penetrating characteristics of pile and the calculation achieved is useful in engineering.

关 键 词：人工神经网络 静力压力桩 沉桩阻力 静力触探试验 

分 类 号：TU473.1[建筑科学—结构工程]