Search

• A blowout failure mechanism of elliptic tunnel faces is developed. • Blowout face pressure are sensitive to the shape of elliptical tunnel faces. • Spatially rotated anisotropy varying strength parameters are considered. • A case study is presented to demonstrate the practical use of the proposed model. Because of high space utilization, elliptical cross-section sometimes is adopted in the design of urban pedestrian underpasses, underground common ditches and subway tunnels. Nevertheless, theoretical models on the blowout failure analysis of shallow elliptical tunnel faces are quite rare. Within the framework of the kinematical approach of limit analysis, deterministic and probabilistic blowout analyses of a shallow elliptical tunnel in frictional-cohesive soils are performed in this study. With the help of the spatial discretization technique, an improved blowout failure mechanism for shallow elliptical tunnel faces is generated. Compared with previous analytical or semi-analytical ones, such an improved failure mechanism allows variable frictional angles of soils to be readily accounted for. The critical blowout face pressure is determined based on the work rate balance equation and a hybrid global optimization algorism. The deterministic model is validated by numerical simulations, existing experimental results and theoretical solutions, in which the effect of the geometrical parameter of elliptical tunnel faces and the less computational cost over numerical simulations are emphasized. Deterministic parametric analyses are conducted to investigate the influences of several significant parameters in the form of charts. Probabilistic blowout analyses of tunnel faces excavated in spatially rotated anisotropy varying soils are then conducted via a combination of the subset simulation, the random field theory and the Markov Chain Monte Carlo method, and the influence of characteristic parameters of the random field are examined. In the end, a case study is presented to demonstrate the practical use of the proposed model. [ABSTRACT FROM AUTHOR]