Akademik Veri Yönetim Sistemi
JOURNAL OF POLYTECHNIC-POLITEKNIK DERGISI, 2024 (ESCI)
Seepage in embankment dams is defined as the continuous and uncontrollable movement of water from the upstream to the downstream side of the dam. During design phase, it is aimed to keep this movement below a certain threshold. For this aim, numerous experimental and numerical studies have been conducted to accurately predict seepage. In recent years, probabilistic analyses incorporating variations in soil properties have been increasingly employed, particularly with the aim of reducing uncertainties in calculations. Finite element methods are frequently employed in these probabilistic analyses. In this study, probabilistic seepage analyses were conducted using the finite element method, considering uncertainties in the permeability values of the core material. Two hypothetical embankment dams with different core slopes were designed for the study. Initially, the statistical parameters of permeability values for the clay core were determined, and permeabilities generated through Monte Carlo simulations were employed as inputs for PLAXIS 2D analyses. Results from both models with the same mean permeability were compared with each other and with the deterministically obtained results. Generally, the study reveals lower seepage values through probabilistic analysis compared to those obtained with deterministic methods. Furthermore, the impact of the assumed probabilistic distribution parameters of the soil variability on the final outcome was explicitly illustrated.