欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对地铁工程盾构法施工引起的邻近建筑物沉降问题,运用最小二乘法和BP神经网络法对现场监测的数据进行拟合、预报,并与之对比.研究结果表明：同一数值拟合方法下不同次数的曲线拟合值差别较大,即拟合次数越多,拟合预测值越准确;BP神经网络法表现出较高的拟合度和预报精准性.BP神经网络的数据拟合相较于传统方法计算便捷,可提高地铁施工过程中邻近建筑物沉降的监测效率.
英文摘要：
In view of the settlement of adjacent buildings caused by shield tunneling,the least square method and BP neural network method are used to fit and predict the data of the field monitoring. The re- sults show that the curve fitting values of different times under the same numerical fitting method vary greatly,that is, the more times of fitting, the more accurate the prediction value is; BP neural network method performs a higher degree of fitting and prediction accuracy. Compared with the traditional method, the numerical simulation of BP neural network is convenient and can improve the efficiency of the settle- ment monitoring of adjacent buildings in the subway construction.
同期刊论文项目
同项目期刊论文
期刊信息
位置：