Prediction of the Stability of Various Tunnel Shapes Based on Hoek–Brown Failure Criterion Using Artificial Neural Network (ANN)

In this paper, artificial neural network (ANN) models are presented in order to enable a prompt assessment of the stability factor tunnels rock masses based on Hoek–Brown (HB) failure criterion. Importantly, safety is one serious concerns for constructing and requires reliable accurate analysis. However, it challenging engineers construct finite element limit analysis (FELA) algorithms with HB criterion tunnel solutions masses. For first time, machine-learning-aided prediction proposed paper. Three different shapes tunnels, i.e., heading tunnel, dual square circular considered. The inputs include four dimensionless parameters including cover-depth ratio, normalized uniaxial compressive strength, geological strength index (GSI), mi parameter. Moreover, more additional parameter namely distance ratio. results present best ANN each shape, providing very predicting