| 盾构施工引起地面沉降的双子神经网络预测 |
| |
| 引用本文： | 李宗梁,银鸽.盾构施工引起地面沉降的双子神经网络预测[J].地下空间与工程学报,2014,10(1):191-200. |
| |
| 作者姓名： | 李宗梁 银鸽 |
| |
| 作者单位： | 1.浙大网新集团有限公司,杭州 310004;2.浙江大学建筑工程学院,杭州 310058 |
| |
| 基金项目： | 杭州庆春路过江隧道科研资助项目 |
| |
| 摘 要： | 通过对杭州市庆春路钱塘江隧道的地面沉降值的监测数据进行分析,提出了采用双子网络来进行盾构法隧道地面沉降预测的可行性。采用杭州市庆春路钱塘江隧道的地面沉降值的监测数据来分别训练具有双子网络和无子网络的人工神经网络,并利用杭州市庆春路钱塘江隧道的地面沉降值的监测数据本身以及杭州市沿江大道运河隧道的地面沉降监测数据作为检测数据,进行了四组实验。实验表明,采用了双子网络的人工神经网络相比于无子网络的人工神经网络有着更佳的拟合、预测精度以及更小规模的数据输入量。
|
| 关 键 词： | 神经网络 子网络 盾构法 地面沉降值 |
| 收稿时间： | 2013-09-12 |
Application of Artificial Neural Network with Two Subnets for Predication of Ground Settlement during Shield Construction |
| |
| Li Zongliang,Yin Ge.Application of Artificial Neural Network with Two Subnets for Predication of Ground Settlement during Shield Construction[J].Chinese Journal of Underground Space and Engineering,2014,10(1):191-200. |
| |
| Authors: | Li Zongliang Yin Ge |
| |
| Affiliation: | 1.Insigma Technology Co.,Hangzhou 310004,China;2. College of Civil Engineering and Architecture,Zhejiang University,Hangzhou 310058,China |
| |
| Abstract: | The analysis of ground settlement monitoring data in Qingchun Road Tunnel project in Hangzhou city shows probability of using ANN with two subnets to predict the ground settlement during shield construction. In this research, with the ground settlement measuring data in Qingchun Road Tunnel project as training data, and with ground settlement measuring data in Qingchun Road Tunnel project and in Yanjiang Avenue Tunnel project in Hangzhou as test data, four experiments are conducted. The results of experiments prove the better regression and predicting accuracy and smaller data input quantity of the ANN with subnets than the ANN without subnet. |
| |
| Keywords: | ANN(Artificial Neural Network) subnet shield method ground settlement |
|
| 点击此处可从《地下空间与工程学报》浏览原始摘要信息 |
|
点击此处可从《地下空间与工程学报》下载全文 |
