摘要
以太原市某基坑沉降监测点的实测数据为基础,选择20期数据进行分析;建立具有2个隐含层的BP神经网络模型,以18期数据建模,采用新陈代谢的方式对后2期数据进行逐一预测验证。通过K3观测点和K6观测点的实测数据验证与分析,认为BP神经网络模型预测精度达到工程应用要求。认为该研究所用方法在相似工程变形的数据预测中可以继续应用,具有一定的实际应用价值,有利于变形监测数据预测。
Based on the measured data of a foundation pit settlement monitoring point in Taiyuan,20 periods of data are selected for analysis;a BP neural network model with two hidden layers is established,which is modeled with 18 phases of data,and the latter two phases of data are predicted and verified one by one by means of metabolism.Through the verification and analysis of the measured data of K3 observation point and K6 observation point,it is considered that the prediction accuracy of BP neural network model meets the requirements of engineering application.It is considered that the method used in this study can continue to be applied in the data prediction of similar engineering deformation,has certain practical application value,and has certain reference significance for the prediction of deformation monitoring data.
作者
赵贞
Zhao Zhen(Shanxi Branch of China National Geological Exploration Center of Building Materials Industry,Taiyuan 030031,China)
出处
《山西建筑》
2022年第8期160-162,共3页
Shanxi Architecture
关键词
BP神经网络
深基坑
变形监测
预测
back propagation neural network
deep foundation pit
deformation monitoring
prediction
