首页 | 本学科首页   官方微博 | 高级检索  
     检索      

神经网络在单桩承载力预测中的应用
引用本文:韩胜德,孟春红,韩丽华,胡忠璜.神经网络在单桩承载力预测中的应用[J].水运工程,2007(10):12-15.
作者姓名:韩胜德  孟春红  韩丽华  胡忠璜
作者单位:1. 广东金东海集团大连分公司,辽宁,大连,116011
2. 中国大连石油化工公司,辽宁,大连,116032
3. 大连理工大学,辽宁,大连,116024
摘    要:运用基于BP神经网络的组合预测模型对PHC桩的极限承载力进行预测。分别利用灰色GM(1,1)模型和BP神经网络对桩在荷载作用下的沉降进行估算,然后利用人工神经网络中的BP网络对所得的结果进行组合预测;最后利用Lagrange算法计算桩的极限承载力。计算实例表明,使用该组合预测方法所得的预测结果比单纯使用灰色GM(1,1)模型或神经网络模型所得结果的总体误差要小,因而该方法是可行的、有效的。

关 键 词:灰色系统理论  PHC桩(超高强预应力管桩)  神经网络  极限承载力  组合预测
文章编号:1002-4972(2007)10-0012-04
收稿时间:2007-05-16
修稿时间:2007年5月16日

Application of ANN in Determining Ultimate Bearing Capacity of a Single Pile
HAN Sheng-de,MENG Chun-hong,HAN Li-hua,HU Zhong-huang.Application of ANN in Determining Ultimate Bearing Capacity of a Single Pile[J].Port & Waterway Engineering,2007(10):12-15.
Authors:HAN Sheng-de  MENG Chun-hong  HAN Li-hua  HU Zhong-huang
Institution:1. Dalian Branch of Guang dong Gold tast Sea Greup Co., Dalian 116011, China; 2. China Petroleum Dalian Petrochemical Corporation, Dalian 116032, China; 3. Dalian University of Technology, Dalian 116024, China
Abstract:A joint forecasting model of artificial neural network(ANN) is utilized to estimate the ultimate bearing capacity of PHC pile(pre-stressed pipe pile with super high strength).Firstly,the gray GM(1,1) and the ANN are separately used to estimate the subside of single pile under load.Then the BP neural network is employed to forecast the subside based on the above two estimating results.At last,the ultimate bearing capacity of single pile is got through Lagrange arithmetic.The result shows that the total forecasting error by this method is smaller than by ANN or gray GM(1,1) alone.So this method is effective and feasible.
Keywords:grey system theory  PHC pile(pre-stressed pipe pile with super high strength)  artificial neural network  ultimate bearing capacity  joint forecasting
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号