Prediction of Subsidence during TBM Operation in Mixed-Face Ground Conditions from Realtime Monitoring Data

Abstract

The prediction of settlement during tunneling presents multiple challenges, as such settlement is governed by not only the local geology but also construction methods and practices, such as tunnel boring machine (TBM). To avoid undesirable settlement, engineers must predict the settlement under given conditions. The widely used methods are analytical solutions, empirical solutions, and numerical solutions. Analytical or empirical solutions, however, have limitations, which cannot incorporate the major causes of subsidence, such as unexpected geological conditions and TBM operational issues, among which cutterhead pressure and thrust force-related factors are the most influential. In settlement prediction, to utilize the machine data of TBM, two phases of long short-term memory (LSTM) models are devised. The first LSTM model is designed to capture the features affecting surface settlement. The second model is for the prediction of subsidence against the extracted features. One thing to note is that predicted subsidence is the evolution of settlement along TBM drive rather than its maximum value. The proposed deep-learning models are capable of predicting the subsidence of training and test sets with excellent accuracy, anticipating that it could be an effective tool for real-world tunneling and other underground construction projects.

Prediction method of longitudinal surface settlement caused by double shield tunnelling based on deep learning

“…To shorten the construction period, numerous engineering projects employ the strategy of sequential excavation of two tunnels 42 , which collectively contribute to surface settlement. However, most existing studies primarily focus on the parameters of a single tunnel 43 – 45 , overlooking those of the other tunnel. This approach, despite its cost-saving implications, impacts calculation accuracy due to the potential for increased surface settlement resulting from the excavation and geotechnical actions in dual-line tunnels 46 .…”

Intelligent Analysis of Construction Safety of Large Underground Space Based on Digital Twin

“…Han et al [10] introduced a method of using BIM and 3D laser scanning technology to monitor the deformation of a foundation pit, and took a large foundation pit as an example to verify that the method can monitor the three-dimensional overall deformation of foundation pits efficiently and accurately. In the prediction of surface subsidence caused by mining, Lee et al [11] proposed two models based on a long short-term memory network to capture the characteristics of surface subsidence caused by mining and predict subsidence. The proposed deep learning model can accurately predict the subsidence of the training set and the test set.…”

Reinforcement learning for the face support pressure of tunnel boring machines

“…Some promising research domains for machine learning in tunnelling are the geological prognosis ahead of the face, the interpretation of monitoring results, automation and maintenance [32]. At present, however, research appears to be focussed on the following topics: prediction of TBM operational parameters [34,[39][40][41][42][43][44][45][46], penetration rate [47][48][49][50][51][52][53][54][55][56][57][58][59][60][61][62][63], porewater pressure [64], ground settlement [65][66][67], disc cutter replacement [68][69][70], jamming risk [71,72] and geological classification [73][74][75][76]). Few authors estimated the face support pressure of TBMs with machine learning [35,52].…”