A practical ANN model for predicting the excavation-induced tunnel horizontal displacement in soft soils

Abstract

The objective of this study is to propose an artificial neural network (ANN) model to predict the excavation-induced tunnel horizontal displacement in soft soils. For this purpose, a series of finite element data sets from rigorously verified numerical models were collected to be utilized for the development of the ANN model. The excavation width, the excavation depth, the retaining wall thickness, the ratio of the average shear strength to the vertical effective stress, the ratio of the average unloading/reloading Young’s modulus to the vertical effective stress, the horizontal distance between the tunnel and retaining wall, and the ratio of the buried depth of the tunnel crown to the excavation depth were chosen as the input variables, while the excavation-induced tunnel horizontal displacement was considered as an output variable. The results demonstrated the feasibility of the developed ANN model to predict the excavation-induced tunnel horizontal displacement. The proposed ANN model in this study can be applied to predict the excavation-induced tunnel horizontal displacement in soft soils for practical risk assessment and mitigation decision.