Evaluating the Performance of Geosynthetic Reinforced Soil-Integrated Bridge System (GRS-IBS) under Working Stress Condition

This paper evaluates the performance of geosynthetic reinforced soil-Integrated Bridge System (GRS-IBS) in terms of lateral facing deformation and strain distribution along geosynthetics. Simulations were conducted using 2D PLAXIS program. The hardening model proposed by Schanz et al. [1] was used to simulate the behavior of backfill material; the backfill-reinforcement interface was simulated using Mohr-Coulomb model, and the reinforcement and facing block were simulated using linear elastic models. The numerical model was verified using the results of a case study conducted at Maree Michel GRS-IBS, Louisiana. Parametric study was carried out to investigate the effects of span length, reinforcement spacing, and reinforcement stiffness on the performance of GRS-IBS. The results indicate that span length have significant impact on strain distribution along geosynthetics and lateral facing deformation. The reinforcement stiffness has significant impact on the GRS-IBS behavior up to a certain point, beyond which the effect tends to decrease contradictory to reinforcement spacing that has a consistent relationship between the GRS-IBS behavior and reinforcement spacing. The results also indicate that reinforcement spacing has higher influence on the lateral facing deformation than the reinforcement stiffness for the same reinforcement strength/spacing ratio (Tf/Sv) due to the composite behavior of closely reinforcement spacing.


Introduction
The composite behavior of internally supported reinforced soil, the Geosynthetic Reinforced Soil walls (GRS), has advantages over the traditional concrete walls due to the ease of construction, cost saving, and construction time. In addition to the support of the selfweight of the backfill soil, the GRS walls can support the roadway structures and traffic loads [2][3][4][5][6]. A relatively new use of this system is in bridge application [Geosynthetic Reinforced Soil-Integrated Bridge System (GRS-IBS)], which can help reduce both the bridge construction time and cost [e.g. 7-13] The GRS-IBS usually includes a GRS abutment, bearing bed reinforced zone, GRS integrated approach, and a reinforced soil foundation [14]. The GRS-IBS can be used to integrate the bridge structure with the approaching road to create a jointless bridge system. Two versions of GRS-IBS are defined by the FHWA, one version uses steel girders with either a CIP footing or a precast sill. Another version of GRS-IBS uses adjacent concrete box beams supported directly on the GRS abutment without a concrete footing. Many numerical studies have been conducted on the behavior of a free-standing geosynthetic mechanically stabilized earth (GMSE) walls [e.g. [15][16][17][18][19][20][21][22]. A few numerical studies were conducted recently to evaluate the composite behavior of the GRS-IBS [e.g. [23][24][25][26][27]12].

Numerical model
The two-dimensional finite element program PLAXIS 2D 2016 [28] was used in the current study to evaluate the effect of different parameters on the performance of GRS-IBS. Mesh refinement was first conducted to find the optimum mesh-size where the numerical results are not mesh-size dependent. The dimensions of the model domain were selected far enough to minimize the effect of boundary conditions on the model response. The lateral boundaries were fixed by roller support to prevent the soil movement in the horizontal direction. The bottom of soil foundation was fixed using bin support to prevent the soil from movement in both the horizontal and vertical directions.
The model was successfully developed and used to simulate the behavior of Maree Michel GRS-IBS [7][8][9] during the construction stages and after the bridge was opened to traffic. The Maree Michel Bridge was constructed by the Louisiana Department of Transportation and Development (LA DOTD) in Vermilion Parish in 2016. The total height of the GRS-IBS wall is 3.8 m from the top of the RSF and was divided into 20 layers to simulate the field construction process by using the staged construction mode in PLAXIS 2D 2016, which allows for simulation of construction and excavation processes. A 63 kPa distribution load at the top and bottom and exposed faces of each soil layer was applied during the staged construction process to simulate the soil compaction. This approach is based on the procedure introduced by Dantas [29] to consider the induced stress on the backfill soil due to compaction, which was also adopted later by Ehrlich and Mirmoradi [30], Mirmoradi and Ehrlich [20,31]. The triaxial and large direct shear testing method were conducted to evaluate the strength and stiffness of the backfill materials properties. A total of three triaxial testing were conducted at three different confining pressures of 207, 345, and 483 kPa for a soil specimen size of 15.24 cm diameter and 30.48 cm height. Fig. 1 presents the simulated and measured stress-strain curve for the backfill materials [32]. Six large direct shear tests having with dimension size of 30.48 × 30.48 × 15.24 cm were conducted to evaluate the strength properties and the interface friction angle between the geosynthetic and the backfill/facing block materials. The block/soil interface friction is 27.7° based on a previous study conducted by Ling et al. [33]. A jointless interface between the bridge slab and the integrated approach was simulated based on FHWA [14]. The interface between the bridge and the footing was simulated with a friction coefficient of 0.4 [26]. The tests were conducted under normal stresses of 48.3, 120, and 191.7 kPa, which results in peak stresses of 83.5, 144, and 260 kPa, respectively. The dilation angle was estimated using the following reference [34]; Where: ϕp = peak friction angle = 51°; ϕcr = critical state friction angle = 34°; ψ = dilation angle. In the current study, the configuration of the GRS-IBS numerical model is selected according to the FHWA design criteria recommendation [14]. Figure 2 presents the configuration of the GRS-IBS model that was adopted to perform the parametric study. The height of bridge abutment H was selected with a minimum span length Lspan larger than 7.6 m, the minimum base width Btotal is the greater value of 1.83 m or 0.3H. The width of the reinforced soil footing (RSF) Brsf is equal to Btotal +0.25Btotal, assuming the bridge span to depth ratio = Lspan/D = 24 as reported by Zheng and Fox [27], and the depth of the RSF Drsf is equal to 0.25 Btotal. The setback distance between the back of the face and the footing ab is equal to 0.2 m. The minimum clear space de, the distance from the top of the facing block to the bottom of the superstructure, is equal to 8 cm or 2% of the abutment height, whichever is greater. The width of the beam seat (strip footing in this study) b was selected equal to 1.2 m with a thickness of 0.6 m (note that the minimum width of the beam seat for a span length greater than 7.6 m is 0.77 m and the minimum thickness is 0.2 m). The minimum reinforcement length Lr at the bottom of the bridge abutment should be 0.3H or Btotal, whichever is greater, which increases linearly up to 0.7H. The bearing bed reinforcement zone was extended from the top reinforcement layer for six consecutive layers. The length of the bearing bed reinforcement Lrb is equal to 2ab+b.

Effect of reinforcement stiffness
Five different reinforcement stiffness, EA, were considered and evaluated in this study: 300, 600, 900, 1200, and 1500 kN/m. Fig. 7 presents the strain distribution along the reinforcement at 20, 40, 60, and 80% of the abutment height as measured from the bottom of abutment. Similar to the effect of reinforcement spacing, the reinforcement stiffness affects the magnitude of the strain but does not affect either the shape of the strain distribution or the location of maximum strain. The maximum strain decreases from about 1.3% for a reinforcement stiffness of 300 kN/m to about 0.5% for a reinforcement stiffness of 1500 kN/m. It can be seen that increasing the reinforcement stiffness from 300 kN/m to 900 kN/m has significant effect on the reinforcement strain (e.g., the strain decreases from 1.3% to 0.68% at 0.8 H). However, after that, the effect of reinforcement stiffness tends to decrease (e.g., the strain decreases from 0.6% to 0.5% when the reinforcement stiffness increases from 900 kN/m to 1500 kN/m at 0.8 H). Fig. 8 presents the effect of reinforcement stiffness on the lateral displacement of wall facing. The maximum lateral facing displacement decreases from 37 mm for a reinforcement stiffness of 300 kN/m to 26 mm for a reinforcement stiffness of 1500 kN/m.

Conclusions
Finite element parametric study was conducted to evaluate the performance of the GRS-IBS in terms of lateral deformation and reinforcement strain. The parameters included in this study are the effects of span length, the height of GRS abutment, the reinforcement spacing, and the reinforcement stiffness. Based on the finding of this study, the following conclusions can be made: The results indicate that while the magnitude of reinforcement strain is affected by span length and abutment height, the shape of the strain distribution is not affected. • The reinforcement spacing has significant influence on the strain distribution along the reinforcement and the lateral facing displacement, in which the maximum strain and lateral facing displacement increase with increasing reinforcement spacing. The maximum strain increases from 0.6% for a reinforcement spacing of 0.1 m to 1.4% for a reinforcement spacing of 0.4 m, and the maximum lateral facing displacement increases from 28 mm for a reinforcement spacing of 0.2 m to 42 mm for a reinforcement spacing of 0.4 m. • The reinforcement stiffness has significant influence on the behavior of GRS-IBS in terms of reducing the lateral facing displacement and the magnitude of strain distribution along the reinforcement with increasing stiffness up to a certain point, after which this impact tends to decrease in contrary to the effect of reinforcement spacing, which shows a constant impact on the performance of GRS-IBS.