Enhancing Analytical Methods for Estimating Water Inflow to Tunnels in the Presence of Discontinuity Areas

Abstract

Abstract There are various methods for calculating the inflow of water into excavated tunnels in rock formations, including empirical, analytical, and numerical methods. Analytical equations are widely used due to their simplicity and reliance on assumptions. However, past studies have shown that the measured water inflow into rock tunnels is often lower than the values estimated using analytical equations. Moreover, results obtained using analytical equations are highly dependent on the tunnel geometry and environmental conditions. Hence, this study employed finite element numerical modeling to simulate the effects of various factors, including fault distance from the tunnel, permeability coefficient, fault width, tunnel radius, rock mass permeability, and groundwater level, on the water inflow into the tunnel. The analytical method was then used to estimate the water inflow, and the results were compared with the numerical modeling outputs. Subsequently, modified equations were developed to estimate the water inflow under different conditions, including cases where the fault intersects or does not intersect the tunnel. The correlation between the results obtained from the equations and the numerical modeling outputs was evaluated using R and R2 statistics. The obtained values were within an acceptable range, indicating the validity of the proposed models. Furthermore, the histograms of the residuals for both models showed a good fit. To validate the proposed models, the analytical method and the proposed equations were used to estimate the water inflow into the Amirkabir tunnel, respectively. The results showed that the values obtained using the new equations were closer to the actual values than those obtained using the analytical method. This study highlights the importance of considering various factors in estimating the water inflow into rock tunnels and provides new equations that can improve the accuracy of such estimates.