基于小波神经网络技术的基坑水平位移形变预测  

Deformation prediction of foundation pit horizontal displacement based on wavelet neural network

作　　者：刘永辉 Liu Yonghui(Tianjin Surveying&Designing Institute for Water Transport Engineering,Tianjin 300456,China;Tianjin Research Institute for Water Transport Enginering,M.0.T,Tianjin 300456,China)

机构地区：[1]天津水运工程勘察设计院,天津300456 [2]交通运输部天津水运工程科学研究所,天津300456

出　　处：《山西建筑》2021年第9期54-55,共2页Shanxi Architecture

基　　金：交通运输部天津水运工程科学研究所基金项目:基于高精度自动化监测技术和物探技术的深基坑变形监测新方法的研究与应用(TKS170214)

摘　　要：基坑形变预测有助于工程建设决策,为提高形变预测的精度,提出将形变时序进行小波分解重构后进行神经网络预测,并用实测数据进行验证。结果表明:与BP神经网络预测相比,小波神经网络能取得更好的预测效果。In order to improve the accuracy of deformation prediction,the deformation time series is proposed by wavelet decomposition and reconstruction,and the neural network prediction is carried out. The measured data is used for verification. The results show that compared with direct neural network prediction,wavelet neural network can achieve better prediction results.

关 键 词：形变 预测 小波分解 神经网络 

分 类 号：TU433[建筑科学—岩土工程]