AI-Driven Predictive Seepage Analysis of Gangapur Earthen Dam Using Geo-Studio Software

Abstract

Seepage is an essential element that affects both the stability and functionality of earthen dams, where uncontrolled seepage can potentially result in structural failure. This research investigates the seepage characteristics of the Gangapur earthen dam utilizing Geo-Studio SEEP/W, a software based on the finite element method, to evaluate its precision in relation to conventional empirical methods. Critical factors including the elevation of the phreatic surface, distribution of pore pressure, hydraulic gradient, and rates of seepage were analyzed to assess the stability of the dam. The simulation outcomes demonstrate a strong correlation (R² > 0.95) between the observed and numerical seepage data, validating the dependability of finite element modeling. The seepage discharge fluctuates between 0.019 and 0.038 m³/s, with peak pore pressure reaching 60 kPa, ensuring that seepage forces remain beneath the critical limit. Furthermore, the factor of safety remains above 1.28 under severe conditions, indicating structural integrity. These results confirm the efficacy of numerical seepage analysis, reinforcing its role in the evaluation of dam safety. The study underscores the significance of real-time seepage monitoring and AI-driven predictive modeling for optimizing seepage control strategies and improving the long-term performance of dams.