Stability evaluation of elliptical tunnels in natural clays by integrating FELA and ANN

Abstract

The stability of tunnels in clayey soil is a major concern for underground space technology. Clay has anisotropy in shear strength induced by depositional and sedimentation processes. For the numerical analysis of geotechnical stability problems, the anisotropic undrained shear (AUS) model can account for this anisotropy of clayey soils. In this study, the stability of the elliptical tunnel (stability factor: σs-σt/suc) with varying elliptical shape (width-depth ratio: B/D) placed at different embedment depths (cover-depth ratio: C/D) in clay with different anisotropy (anisotropic strength ratio: re) and varying dimensionless overburden factor (overburden factor: γD/suc) is evaluated using finite element limit analysis and the AUS model. The failure planes are also evaluated for the above variations. Based on the numerical outcome, the artificial neural network (ANN) is utilized to establish the equation for predicting the stability of the elliptical shape tunnel with different shapes (i.e., width-depth ratio), and varying overburden, cover-depth ratio, varying anisotropic strength ratio of clay. The present study results are presented as design charts, tables, and equations so that they can be used in practice.

Keywords

• Anisotropic clay
• Elliptical tunnel
• Finite element limit analysis
• AUS model
• Artificial neural network