摘要
地连墙侧向位移对深基坑施工风险防控至关重要,南方地下轨道工程地质多为软土,且存在土层复杂不均等现象,土体具有明显不确定性,基坑施工存在安全隐患。为提升基坑安全性,文中基于贝叶斯参数更新框架建立数值模型,结合响应面法理论,对基坑地连墙侧向位移进行预测。结果表明,该基坑地连墙侧向位移的预测值与实际观测值基本相似,表明该方法具有可行性;GA-BP神经网络响应面法可作为响应面函数;更新土体参数后的预测值与观测值差异减小,地连墙多点观测的误差更小,说明更新土体参数后预测效果更佳。
The lateral displacement of diaphragm wall is very important to the risk prevention and control during the construction of deep foundation pit.In the south of China,the underground track engineering geology is mostly soft soil,and the soil layer is complex and uneven,which leads to the uncertainty of the soil and the hidden danger in the foundation pit construction.In order to improve the safety of foundation pit,a numerical model based on Bayesian parameter updating frame was established to predict the lateral displacement of diaphragm wall.The results show that the predicted value of lateral displacement is similar to the observed value,and the GA-BP neural network response surface method can be used as response surface function The difference between the predicted value and the observed value decreases after the soil parameters are updated,and the error of the multipoint observation of the diaphragm wall is smaller,which indicates that the prediction effect is better after the soil parameters are updated.
作者
林义港
Lin Yigang(Fujian Construction Engineering Group Co.Ltd.,Fuzhou,Fujian 350000)
出处
《江西建材》
2023年第9期169-171,共3页
Jiangxi Building Materials
关键词
软土
深基坑
不确定性位移
数值模拟
Soft soil
Deep foundation pit
Uncertainty displacement
Numerical simulation
