Prediction model of particle loss based on seepage tests of sediment in water-level-fluctuation zone of reservoir

Particle erosion is one of the important factors in the deformation and failure of sediment in the reservoir slopes during reservoir operation. The determination of the sensitive factors and the degree of loss of party les from reservoir banks is an important prerequisite for studying the stability of reservoir banks. The degree of particle loss has become a key indicator in studying the deformation failure process of reservoir slope sediment. Based on field investigation and statistics, the coarse-grained skeleton of slope sediment was divided into five types. An orthogonal design method was used to design the test specimens. In this study, a multi-valve parallel particle erosion test system was developed, and a preparation method of sediment samples was proposed. The variation of grain positions in the sediment sample is analyzed using digital image processing technology. The sensitivity factor of fine particle loss rate was obtained by the range analysis method. Based on seepage test results, a prediction model of particle loss rate was proposed. The prediction model was verified by comparing it with experimental data from two mounded slope sites, Baijiabao landslide and Kaziwan landslide in the Three Gorges Reservoir area. The results show that the relative density is more sensitive than other factors in affecting the particle loss rate. The particle loss is one of the non-negligible reasons for the sediment strength deterioration of the reservoir water-level-fluctuation zone. The results of the study can provide useful refer-ences for future studies on the macroscopic and microscopic structural characteristics of bank slope sediment during reservoir operation.