基于GA-Bi-LSTM的盾构隧道下穿既有隧道沉降预测模型  

Settlement prediction model of shield tunnel under-crossing existing tunnel based on GA-Bi-LSTM

作　　者：周中[1] 张俊杰 丁昊晖 李繁 ZHOU Zhong;ZHANG Junjie;DING Haohui;LI Fan(School of Civil Engineering,Central South University,Changsha,Hunan 410075,China)

机构地区：[1]中南大学土木工程学院,湖南长沙410075

出　　处：《岩石力学与工程学报》2023年第1期224-234,共11页Chinese Journal of Rock Mechanics and Engineering

基　　金：国家自然科学基金资助项目(50908234);湖南省自然科学基金项目(2020JJ4743)。

摘　　要：为了对盾构隧道下穿既有隧道的安全控制提供支撑,以既有地铁隧道的沉降预测误差为目标,采用遗传算法(GA)对双向长短期记忆网络(Bi-LSTM)结构中的时间序列的长度、隐藏层的单元数、隐藏层层数、LSTM的层数以及dropout进行参数优化,并在综合考虑工程地质参数、空间参数和盾构施工参数的基础上,构建GA-Bi-LSTM既有隧道沉降预测模型。以长沙轨道交通3号线平行下穿长沙轨道交通1号线区间工程为依托,基于该区间内的既有隧道沉降监测值以及对应的输入参数数据,对模型进行训练和测试。研究结果表明:提出的GA-Bi-LSTM模型对既有隧道沉降预测的平均绝对误差(MAE)、均方根(RMSE)、样本回归值(R^(2))分别为0.42,0.45,0.90,平均相对误差仅为10.78%,相较于BP,SVM,LSTM,Bi-LSTM神经网络模型拥有更好的预测精度,说明该模型具有较好的可靠性和实用性,可为新建隧道下穿既有隧道的沉降预测提供一种新的思路和方法。In order to provide support for the safety control of shield tunnel under-crossing the existing tunnel,genetic algorithm(GA)is used to optimize the structural parameters of Bi-directional Long Short-Term Memory(BiLSTM),including the time series,units in hidden layer,hidden layers,Bi-LSTM layers,and dropout.And then,the existing tunnel settlement prediction model,named GA-Bi-LSTM,is constructed by comprehensively considering engineering geological parameters,spatial parameters,and shield construction parameters.Based on the settlement monitoring values and the corresponding construction parameters of the section project of Changsha Rail Transit Line 3 crossing Changsha Rail Transit Line 1 in parallel,the model is trained and tested.The results show that the mean absolute error(MAE),root mean square error(RMSE)and sample regression value(R^(2))of the GA-Bi-LSTM model are 0.42,0.45 and 0.90 respectively,and the average relative error is only 10.78%.Compared with BP,SVM,LSTM and Bi-LSTM models,GA-Bi-LSTM model has better prediction accuracy,indicating that the model has better reliability and practicability,which can provide a new idea and method for the settlement prediction of the new tunnel under-crossing the existing tunnel.

关 键 词：隧道工程 盾构隧道 神经网络 遗传算法 预测模型 沉降量 

分 类 号：U45[建筑科学—桥梁与隧道工程]