摘要
为使监测模型深入揭示深基坑在开挖期间的变形规律,从土体遗传蠕变机理出发,提出等效开挖深度概念,将变形影响因子构造为考虑开挖深度的瞬时变形影响因子和考虑蠕变效应的历史变形影响因子.利用径向基函数神经网络的强大的非线性映射能力,以已有的实测数据为训练样本,构造相应的输入因子,建立了深基坑变形的监测模型,可实现对后期开挖的深基坑变形的非线性预测.实例验证表明,该模型效果好、有利于对开挖作用下的深基坑变形进行监测分析和预测,为保障深基坑变形安全提供了有力工具.
Horizontal deformation of foundation pit during excavation is impacted by the excavation process. In order to express the regulation and characteristic information of displacement during excavation period of foundation pit, a method based on genetic creep theory and project practice was presented. The equal effect excavation depth was defined. The input factors were formed considering the instantaneous deformation and history deformation. Radial basis function （RBF） was used to establish the monitoring model which describes the cause-effect relationship between deformation and excavation. After procuring the input factors, the model structure was obtained. The displacement in the late excavation stages can be forecast by them in this way with the consideration of creep and excavation. Examples were given with survey and table data. They show that this kind of artificial neural network monitoring model is very good to be used to imitate and predict the deformation of deep foundation pit during excavation work period.
出处
《上海交通大学学报》
EI
CAS
CSCD
北大核心
2009年第6期990-993,共4页
Journal of Shanghai Jiaotong University
基金
国家自然科学基金(50609014)资助项目
上海市重点学科建设项目(B208)资助
关键词
深基坑
监测模型
等效开挖深度
开挖
蠕变
径向基函数神经网络
deep foundation pit
monitoring model
equal-effect-excavation-depth
excavation
creep radial basis function artificial neural network
