基于多实例和DNN的桩基承载力与沉降预测模型构建  

Construction of the Pile Foundation Bearing Capacity and Sedimentation Prediction Model Based on Multiple Living Examples and DNN

作　　者：陈振宇 Chen Zhenyu(Fujian Geotechnical Engineering Survey and Research Institute Co.,Ltd.,Fuzhou 350108,China)

机构地区：[1]福建岩土工程勘察研究院有限公司,福州350108

出　　处：《办公自动化》2024年第9期90-92,共3页Office Informatization

摘　　要：在大型工业厂房、交通桥梁、高层建筑等城建工程中,桩基础应用非常普遍,桩基承载力与沉降直接影响城建工程桩基的安全性与可靠性。基于深度神经网络(DNN)的桩基承载力与沉降预测对提升桩基施工安全性具有重要意义。依托湛江组黏土中桩基承载力与沉降时效性模型试验实例,结合DNN构建桩基承载力与沉降预测模型,经过对数据样本的预处理、网络结构的设计、神经网络的训练等,探究该预测模型对桩基极限承载力与沉降的预测效果。整体上,该模型的预测效果能满足要求,建议增加训练样本提升沉降时效性预测效果。The pile foundation is widely used in urban construction projects such as large industrial plants,traffic bridges and high-rise buildings.The bearing capacity and sedimentation of pile foundation directly affect the safety and reliability of pile foundation in urban construction projects.The prediction of pile bearing capacity and sedimen-tation based on Deep Neural Network(DNN)is of great significance to improve the safety of pile foundation con-struction.Based on the experimental example of pile bearing capacity and sedimentation timeliness model in Zhan-jiang formation clay,the prediction model of pile bearing capacity and sedimentation was built with DNN.After data sample pretreatment,network structure design,neural network training,etc.,the prediction effect of this prediction model on ultimate bearing capacity and sedimentation of pile foundation was explored.On the whole,the prediction effect of this model can meet the requirements,and it is suggested to add training samples to improve the prediction effect of sedimentation timeliness.

关 键 词：工程实例 DNN 桩基承载力 沉降 预测模型 

分 类 号：TU753.3[建筑科学—建筑技术科学]