Three-dimensional stochastic geological modeling for probabilistic stability analysis of a circular tunnel face

This study is devoted to the probabilistic stability analysis of a circular tunnel face incorporating a three-dimensional geological model which is composed of two correlated random fields representing the spatial variability of soil shear strength parameters. The Fast Fourier Transform-Moving Average is adopted for random field generations which takes advantage of discrete Fourier transforms to achieve an easy and time-saving calculation over a large model size. By matching the geological units and failure mechanism elements according to their spatial locations, the employed discretization-based failure mechanism of tunnel face establishes the connection between the stochastic geological model and the mechanical solution model in the framework of limit analysis. By running the Monte Carlo simulation, the failure probability of tunnel face can be obtained for tunnel designs. For the purpose of validation, the proposed method is compared with the numerical approach of FLAC3D by performing deterministic analyses based on the same geological model. Next, parametric analysis is presented to show the influences of autocorrelation length, cross-correlation coefficient, and stratigraphic dip on the failure probability of tunnel face. Finally, a case study considering the failure probability of tunnel face in multilayer soils is discussed to give an example of the practical use of the proposed approach. Conclusively, this method can well simulate the spatial variability of soil properties and the distribution of soil layers; the calculated results can give some guidance for practical tunnel designs.

Automated summary

This study focuses on the probabilistic stability analysis of a circular tunnel face using a three-dimensional geological model with two correlated random fields. The method adopted FFT-MA for random field generation which proved to be efficient and accurate in predicting the failure probability of the tunnel face. Parametric analysis showed the influences of autocorrelation length, cross-correlation coefficient, and stratigraphic dip on the failure probability, providing valuable insights for practical tunnel designs.