路基工程Issue(6)：24-30,7.DOI:10.13379/j.issn.1003-8825.202307017
隧道开挖诱发深基坑变形的LSTM预测模型
LSTM Predictive Model of Deep Foundation Pit Deformation Due to Tunnel Excavation
摘要
Abstract
Aimming at the peoblem of deep foundation pit deformation prediction during tunnel excavation,a prediction method based on Long Short-Term Memory(LSTM)network is proposed.The deformation of deep foundation pits is predicted by collecting relevant data from a real tunnel project and using the time series modeling ability of the LSTM model.To verify the effectiveness of the proposed method,it compares the performance of the LSTM model with traditional prediction methods such as Back Propagation(BP)neural network and Support Vector Machine(SVM)in predicting deep foundation pit deformation caused by tunnel excavation.The results show that the LSTM-based prediction model has higher accuracy and stability than other prediction methods and can effectively predict the deformation of deep foundation pits caused by tunnel excavation.关键词
基坑变形监测/神经网络模型/长短时记忆网络LSTM/预测模型Key words
monitoring of foundation pit deformation/neural network model/long short-term memory network(LSTM)/predictive model分类
交通工程引用本文复制引用
于海涛,丁美跃,林沛元..隧道开挖诱发深基坑变形的LSTM预测模型[J].路基工程,2023,(6):24-30,7.基金项目
国家自然科学基金项目(52008408) （52008408）
中央高校基本业务经费(22hytd06) （22hytd06）
