Maximum surface settlement prediction in EPB TBM tunneling using soft computing techniques

Abstract

Earth pressure balance (EPB) TBMs are commonly used for soft ground tunneling in urban areas. In metro tunnels’ excavation, designing a comprehensive monitoring system to control surface settlement is essential to prevent damage to surface structures. The present study aims to develop new prediction models to estimate the ground surface settlement using two soft computing techniques, SVM and ANN-MLP, and a multiple variable regression model to develop the new empirical formulas. The TBM operational parameters collected from the Tehran metro line 6, South extension (TML6-SE) project have been applied to confirm the provided models. In the data analysis process, the relationships between various parameters (torque index, thrust index, and earth pressure) and the ground surface settlement are investigated. Moreover, several statistical evaluation criteria are implemented to evaluate the performance of the developed models. The results show that the predicted values are in good agreement with the real data. The results can be used for similar ground and TBM tunneling conditions.