Neural Network-Based Prediction Model for the Stability of Unlined Elliptical Tunnels in Cohesive-Frictional Soils

The scheme for accurate and reliable predictions of tunnel stability based on an artificial aeural network (ANN) is presented in this study. Plastic solutions the unlined elliptical tunnels sands are first derived by using numerical upper-bound (UB) lower-bound (LB) finite element limit analysis (FELA). These later used as training dataset ANN model. Note that there four input dimensionless parameters, including overburden factor γD/c′, cover–depth ratio C/D, width–depth B/D, soil friction angle ϕ. impacts these parameters σs/c′ shallow comprehensively examined. Some failure mechanisms carried out to demonstrate effects all parameters. will reliably accurately provide a safety assessment tunnels.