Research Progress on Failure Probability Analysis of Earth-Rockfill Dams Based on Random Field Models

The physical and mechanical parameters of rock and soil inside earth-rockfill dams have strong spatial variability. Random field models can accurately model and analyze failure probability through scientific methods, which helps to better evaluate dam seepage, stability, and seismic performance. This paper reviewed relevant theories of random field models, summarized commonly used covariance and autocorrelation functions in previous research, introduced seven methods for generating random field models and their advantages and disadvantages, and explained the meaning of random field statistical characteristics. This paper summarized domestic and international application examples of random field models in failure probability analysis of earth-rockfill dams from four aspects: seepage, stability, static, and dynamic analysis, including obtaining more accurate seepage, stability, and static dynamic calculation results, analyzing the sensitivity of material parameters of earth-rockfill dams, and playing an important role in parameter inversion problems and establishing dam monitoring models. Future research includes considering non-Darcy flow, 3D fine modeling, and applying new sampling techniques. The relevant achievements can provide a reference for the failure probability analysis of earth-rockfill dams based on random field models.