773
Numerical Evaluation of the Behavior of Reinforced Soil Retaining Walls
Simulation numérique du comportement de murs de soutènement en sol renforcé
Mirmoradi S.H., Ehrlich M.
Dept. of Civil Engineering, COPPE, Federal University of Rio de Janeiro, UFRJ, RJ 21945-970, Brazil
ABSTRACT: In this article, the behavior of reinforced soil walls was studied by performing a numerical analysis using the finite
element method. The numerical approach was validated with the results of a wrapped-faced full-scale reinforced soil wall. In addition,
parametric studies were carried out with different combinations of: facing type, reinforcement stiffness, compaction efforts, and shear
resistance parameters of the backfill soil. Results indicated that for depths below which the vertical induced compaction stresses are
less than the overburden stress, the maximum tension in the reinforcements is the same, irrespective of the values of compaction
effort. However, for lower depths, the tension in the reinforcements increased with the induced compaction stress. In addition, for the
block facing wall, the maximum tension in the reinforcement occurred near the mid-height of the wall. However, for the wrapped
faced wall, the maximum values occurred close to the bottom of the wall. An increase of reinforcement stiffness led to greater values
of tension in the reinforcement for both wrapped and block faced walls. Moreover, an increase of backfill soil shear resistance led to
lower values of tension in the reinforcements.
RÉSUMÉ : Dans cet article, le comportement des murs en terre armée a été étudié en effectuant une analyse numérique utilisant la
méthode des éléments finis. L'approche numérique a été effectuée en se basant sur les résultats d'un mur en terre armée à l’échelle
réelle avec un parement enrobé d’une nappe géosynthétique. De plus, des études paramétriques ont été réalisées avec différentes
combinaisons de type de parement, de rigidité d’armature, d’efforts de compactage et de paramètres de résistance au cisaillement du
remblai. Les résultats ont indiqué que, pour des profondeurs au-dessous desquelles la contrainte verticale de compactage était
inférieure à la contrainte de surcharge, la tension maximale dans l’armature était la même, quelles que soient les valeurs de l'effort de
compactage. Toutefois, pour des profondeurs inférieures, la tension dans l’armature augmentait avec la contrainte de compaction. De
plus, pour le parement mural en bloc, la tension maximale dans l'armature s'est produite à environ mi-hauteur du mur. Cependant,
pour le parement mural enrobé par une nappe géosynthétique, les valeurs maximales se sont produites près du pied du mur. Une
augmentation de la rigidité de l’armature a conduit à de plus grandes valeurs de tension dans l'armature, aussi bien pour le parement
mural en bloc que pour le parement mural enrobé par une nappe géosynthétique. En outre, l'augmentation de la résistance au
cisaillement du remblai a entraîné une baisse des valeurs de tension dans l’armature.
KEYWORDS: Numerical modeling ; Reinforced soil ;Walls ; Compaction effort ; Facing stiffness ; Reinforcement stiffness ;
1 INTRODUCTION
The behavior of reinforced soil was evaluated using finite
element method in the middle 70s (e.g., Romstad et al., 1976).
In recent decades, several numerical analyze using the codes of
the finite element method or finite difference method have been
performed to consider the different geometry and parameters of
GRS walls. Examples are reported by Ling and Leshchinsky
(2003), Hatami and Bathurst (2006), and Guler et al. (2007),
among others.
The purpose of the present study is the numerical evaluation
of the behavior of reinforced soil retaining walls using the finite
element method. The numerical analysis was carried out using
the PLAXIS 2D computer program. The modeling was
validated with the results of a full-scale reinforced soil wall
experiment performed at the Geotechnical Laboratory of
COPPE/UFRJ. Parametric studies were carried out with
different combinations of: facing type, reinforcement stiffness,
compaction efforts, and shear resistance parameters of the
backfill soil.
2 MODEL VALIDATION
The finite element program PLAXIS (Brinkgreve and Vermeer,
2002) was used for the numerical evaluation of the compaction
effect on the behavior of reinforced soil walls. Full-scale
reinforced soil wall modeling, performed at the Geotechnical
Laboratory of COPPE/UFRJ, was used for validation of the
performed analyzes.
The physical model used in this study simulated the behavior
of a 6.8 m high wrapped faced wall (considering the surcharge
load values up to 100 kPa) representing a portion of the
prototype (see Fig. 1). The model wall was 1.4 m high with a
facing inclination of 6° to the vertical. The length and the
vertical spacing of the geogrid were 2.12 m and 0.4 m,
respectively. The value of axial reinforcement stiffness was
equal to 600 kN/m. The model wall was constructed in seven
soil layers, each 0.2 m thick. Layers were compacted by using
both a light vibrating plate (Dynapac LF 81) and a vibratory
tamper (Dynapac LC 71-ET). Equivalent vertical induced
stresses due to soil compaction of 8.0 kPa for the vibrating plate
and 63 kPa for the vibratory tamper were determined. The soil
unit weight after compaction was 21 kN/m
3
. The soil friction
angles, determined by triaxial and plane strain tests, were 42°
and 50°, respectively. Tensions were monitored in the
reinforcements numbered 2, 3, and 4 (see Fig. 1). Load cells
were installed at four points along the reinforcement. The reader
is directed to the paper by Ehrlich et al. (2012) for additional
information about the construction process, the evaluation of the
induced vertical stress due to soil compaction, and the
instrumentation for the performed physical modeling.
