Instability Mechanisms and Reliability Analysis of Bank Slope under Periodic Water Level Fluctuations in Reservoir Areas

The periodic fluctuations of water level in the reservoir area of Baihetan Hydropower Station are prone to inducing bank collapse disasters, and systematic study of the full cycle evolution characteristics of water level rising-stationary-declining phases and the impact of rise and fall rates on reliability analysis is crucial for accurately assessing slope instability risks. Taking the reservoir bank at Xinglong Village section of the Baihetan Hydropower Station as a case study, based on field investigation and reservoir water level data, a corresponding fluid-solid coupled numerical model was established using FLAC3D numerical software. Combined with Karhunen-Loeve expansion of random field theory and Monte Carlo simulation, a stability analysis of the bank slope under periodic water level fluctuations was conducted. The results show that the reservoir bank stability follows a pattern of increasing stability and decreasing risk, and the deformation is mainly concentrated at the contact surface between the gravel mixed soil and the bedrock. A positive correlation exists between the water level fluctuation rate and the failure probability. When the fluctuation rate increased from 0.3 m/d to 1.1 m/d, the failure probability surged from 12.3% to 70.1%. Furthermore, the spatial variability of soil parameters significantly affects the evaluation results of slope stability, and neglecting this characteristic leads to an overestimation of slope stability. The findings reveal the instability mechanisms of the reservoir bank under different rates of periodic water level fluctuations, providing theoretical support for early warning of bank slope disasters and the formulation of water level scheduling schemes in reservoir areas.