Intelligent Prediction of Maximum Ground Settlement Induced by EPB Shield Tunneling Using Automated Machine Learning Techniques

Predicting the maximum ground subsidence (Smax) in construction of soil pressure balanced shield tunnel, particularly on soft foundation soils, is essential for safe operation and to minimize possible risk damage urban areas. Although some research has been done, this issue not solved because its complexity many other influencing factors. Due increasing accuracy machine learning(ML) predicting surface deformation tunneling development automated learning(AutoML) technology. In study, different ML prediction models were constructed using an open source AutoML framework. The model was trained by dataset, which contains 14 input parameters output (i.e., Smax). Different frameworks employed compare their validities efficiencies. performance estimated contrasting parameters, including root mean square error (RMSE), absolute (MAE) determinant coefficient (R2).With a determination (R2) 0.808, MAE 3.7, RMSE 5.2 testing best i.e., extra tree regressor showed better performance, proving that our advantages Smax. Furthermore, SHAP analysis reveal type (ST), torque (To), cover depth (H), groundwater level (GW), deviation have significant effect Smax compared inputs.