Seepage Analysis of Railway Earthen Embankment in Mokama Using RFEM Considering Random Hydraulic Conductivity Field

Abstract

In this study, seepage through the body of an earthen embankment for construction of railway lines in Mokama in Bihar is estimated using finite element method. Seepage of water through the body of earthen embankment adversely affects its stability. In order to check the reliability of the constructed railway embankment stretching for a long distance, Random Finite Element Method (RFEM) originally developed by Fenton and Griffiths (Fenton and Griffiths, 1996) is used. The influence of variability in spatially distributed hydraulic conductivity/permeability field of the embankment is investigated by carrying out seepage analysis. Local Averaging Subdivision (LAS) technique is used to build the random field of hydraulic conductivity of soil domain under consideration. Afterwards, Monte Carlo simulation is performed for 1000 such random fields. The mean of total flow rate $$\left( {m_{Q} } \right)$$ , mean of logarithmic of total flow rate $$\left( {m_{\ln Q} } \right)$$ and variance of logarithmic of total flow rate $$\left( {s_{\ln Q}^{2} } \right)$$ from seepage analyses for 1000 realizations of random field are reported. The results show that as the correlation length decreases, the output parameters $$m_{Q}$$ , $$m_{\ln Q}$$ and $$s_{\ln Q}^{2}$$ follow a decreasing trend and vice versa. These variations are studied for the embankment with and without drain. The effect of drain on the mean hydraulic conductivity and hydraulic gradient standard deviation of the embankment are investigated in which the embankment with drains is found safer than embankment without drain.