Numerical Analysis of Pile-Supported Geogrid-Reinforced Embankments on Soft Grounds

Abstract

In recent decades, the column supported embankments are often constructed at places where soft clay exists within a considerable depth and the construction of roads or rail roads or bridge approach roads is of great demand due to rapid increase of industrialization and urbanization. The column supported embankment has many advantages over the other conventional consolidation based techniques. Such as, these embankments can be constructed at a stretch without prolonged time delay and the embankment loads are directly transferred to the hard strata through piles. Geosynthetics has several advantages for improving the soft grounds, among them providing geogrids as basal-reinforcement below the embankments constructed over soft subsoils of shallow depth is one of the well-known technique. The basal- reinforcements can also be provided above the piles instead of pile caps or raft above piles. The geosynthetics can also be provided in the embankment body to steepen the embankment side slopes. The response of these basal or body-reinforced embankments with or without pile supports under static loading conditions is well-addressed in literature. Most of the studies on dynamic response of these embankments considered cyclic loads or sinusoidal loads to represent traffic loading. Though there are studies available on the seismic response of these embankments, the response of these embankments considering full 3-Dimensional finite element model subjected to time- history loading of different earthquakes is not yet addressed. Hence, in the present study both static and seismic response of basal or body reinforced embankments with and without pile supports are studied using 3-dimensional finite element analysis. In the first part of the study, the response of basal geogrid-reinforced pile- supported embankments subjected to self-weight and traffic load are studied using 3- dimensional finite element models. The influence of various parameters such as, embankment height, geogrid tensile modulus, pile length, pile type and pile spacing are studied. Based on the results of numerical analysis, the modifications to the soil arching coefficient (Cc) including the effect of pile length and pile spacing are proposed and compared with the existing analytical equations. Crest settlements, toe lateral displacements, differential settlements at crest, stress distribution ratio, lateral stress viii distribution ratio and coefficient of lateral pressure along embankment height were considered to analyse the response of these embankments. The analysis of results indicates that, the end-bearing pile supported embankments performs better than floating pile supported embankments in terms of settlements, differential settlements and lateral displacements even at larger pile spacing. The addition of basal geogrid could further reduce the settlements and lateral displacements in the embankment. The analytical equation for Cc proposed based on the 3-dimensional finite element analysis results considered the effect of pile spacing, which the earlier methods did not consider. Hence the proposed analytical equation could able to give the more accurate results of pile loads than the existing methods. The crest centre settlements were further increased by the addition of traffic load. These basal geogrid-reinforced pile-supported embankments should stand safe during disastrous situations like earthquakes. Hence the second part of the study analyses the seismic response of basal geogrid-reinforced pile-supported embankments subjected to seismic excitations. Time-history analysis was performed on the 3- Dimensional finite element models of basal geogrid-reinforced pile-supported embankments. The seismic response of embankment in terms of vertical and lateral displacements, differential settlements, vertical and lateral stress distribution on pile and the foundation soil between piles, amplification coefficient, lateral earth pressure along the embankment height and the pore water pressure are studied by considering the height of embankment, side slope of embankment, basal geogrid tensile modulus, length of pile, spacing of pile and type of pile. The analysis of results shows that the embankment height is an important parameter to consider in the seismic design of basal geogrid reinforcement. 4 m high embankment experiences very less differential settlements caused by seismic excitations among the different embankment heights considered. About 8 % reduction of toe lateral displacements are observed by the addition of basal geogrid. But the embankment with pile supports shows a reduction of 40.8 % and the combination of both pile supports and basal geogrid could reduce 46 % of toe lateral displacements. Addition of basal geogrid increases both vertical and lateral stresses on piles due to seismic excitations. The variation of coefficient of lateral pressure along the embankment elevation is random for the considered parameters, this indicates that the formation of soil arching in a geogrid reinforced pile supported ix embankment subjected to seismic loading is not uniform like in the case of self-weight analysis. Basal geosynthetic-reinforcements are the most commonly used ground improvement technique for the embankments constructed over shallow depth soft clays. The width of the basal-reinforcement provided should be adequate to withstand lateral sliding, rotational failure and excessive settlements under both static and seismic loading conditions. In the third part of the study, an attempt has been made to study the optimum width of basal geosynthetic-reinforcement subjected to both static and seismic loading conditions. Finite element models of basal geosynthetic-reinforced embankments including the effect of embankment height, embankment side slope, tensile modulus of geosynthetic, number of layers of geosynthetic, stiffness of embankment fill, stiffness of foundation soil and intensity of seismic loading were studied. Based on the results of crest settlements, toe lateral displacements and lateral displacements at the crest, the required width and tensile modulus of basal geogrid were identified. Basal geogrid having a minimum tensile modulus of 500 kN/m with a width equal to the base width (B) of embankment is found to be sufficient to reduce settlements at places where static loading is predominant or in low seismic regions. Basal geogrid of width equal to ‘B+H’ having tensile modulus of 4000 kN/m is recommended to reduce the lateral displacements in embankments at active seismic regions. Further reduction of about 6 % in lateral displacements are seen by providing 4 layers of basal geogrid with a total tensile modulus equal to 4000 kN/m. The geosynthetics are also used as embankment body-reinforcements to steepen the embankment side slopes. These slopes are stable under static loading conditions but, under seismic loading conditions, repairable damages or sometimes complete failure of slopes may occur. Hence the present study is also extended to analyse the seismic response of body-reinforced embankments considering the effect of embankment side slope and foundation soil stiffness using finite element analysis. From the analysis it is observed that, in unreinforced embankments the face lateral displacements increase as the steepness of slope increases and the embankment above soft soil displaces more than the embankment on stiff soil.