Print Abstract


GeoAmericas - 2020
Abstract: 361-1

361-1

Numerical modeling of aggregate-geogrid composite behavior for pavement applications using Discrete Element Method

Authors:
Rodrigo Borela1, Sangy Hanumasagar1, Mark Wayne2, David Frost1
1 GT - Georgia Institute of Technology, 2 TC - Tensar Corp

Abstract:

The use of geogrids in road construction has been shown to augment service life and reduce aggregate consumption in pavement base layers (Montanelli et al., 1997). The enhanced performance arises from a combination of lateral confinement of aggregate particles within geogrid apertures and improved aggregate interlocking (Giroud et al., 1985). The properties of the aggregate-geogrid composite systems are governed by several factors including material gradation, aggregate morphological properties, geogrid stiffness, aperture shape and size (Carroll and Haas, 1987; Milligan et al., 1989; Giroud and Han, 2004). In the present study, the effects of these factors on the cyclic-loading behavior of geogrid-stabilized base layers are examined numerically. A series of three-dimensional models were developed using the discrete element method (DEM), consisting of subjecting different specimens to cyclic loading by a cylindrical piston. To capture an adequate range of composite behavior, 12 cases were simulated: 4 stabilization conditions (non-stabilized, triaxial geogrid, biaxial geogrid and full lateral confinement) over 3 aggregate-to-rib size ratios. Aggregates were modeled as monosized spheres interacting via a rolling resistance elasto-frictional contact law (Kozicki et al., 2008), and geogrids followed a deformable element formulation (Effeindzourou et al., 2016). Both components had their micro-parameters independently calibrated using an optimization technique (Borela et al., 2019) to match behavior of real materials. The stabilization cases proposed in this study offer a new framework for comparing aggregate-geogrid systems based on rutting performance as follows. The upper and lower bounds for rutting are established by the non-stabilized and the full lateral confinement cases respectively. In the latter case, horizontal motion is fully constrained on a thin layer of aggregate elements in the specimen while maintaining their free vertical movement. This represents the lower bound for rutting, equivalent to an idealized aggregate-geogrid system where particles near the geogrid are completely laterally restrained.  Establishing these bounds on rutting performance enables an improved assessment of the stabilizing influence of geogrids in aggregate systems. Additionally, the emerging granular phenomena was investigated, providing insights in how the aggregates engage with the geogrids to stabilize the specimen and reduce rutting. Results show an optimum aggregate-to-grid size ratio, indicating that interlocking with the geogrid can be reduced when particles are either too small or large relative to the geogrid openings. Finally, the shape of the geogrid was observed to play an important role in redistributing the load on the granular matrix, effectively changing the degree of rutting reduction.

Keywords:
 geogrid, numerical modeling, discrete element method, aggregate