IFCEE 2015 © ASCE 2015
Numerical Study of Deformation Behavior for a Geosynthetic-Reinforced Soil Bridge Abutment under Static Loading Yewei Zheng1, S.M. ASCE, Patrick J. Fox2, Ph.D., P.E., F. ASCE, and P. Benson Shing3, Ph.D., M. ASCE 1
Graduate Research Assistant, Department of Structural Engineering, University of California, San
Diego, La Jolla, CA 92093; Email:
[email protected] 2
Professor, Department of Structural Engineering, University of California, San Diego, La Jolla, CA
92093; Email:
[email protected] 3
Professor, Department of Structural Engineering, University of California, San Diego, La Jolla, CA
92093; Email:
[email protected]
ABSTRACT: The geosynthetic-reinforced soil (GRS) bridge abutment is a new technology that has many advantages over traditional abutment designs, including cost savings, relatively easy and fast construction, and good performance with regard to differential settlements. The GRS bridge abutment is a complex system that includes a lower GRS wall, bridge seat, and upper GRS wall, with the bridge superstructure load applied directly to the backfill for the lower GRS wall. This paper presents numerical simulations of the static response of a typical GRS bridge abutment during construction and service. The numerical simulations include soil-reinforcement, soil-block, block-block, and soil-bridge seat interactions. Analyses were performed in stages to simulate the abutment construction process. A uniform surcharge load was applied on the bridge deck and approach roadway to simulate traffic loads during service. Results for construction and in-service conditions are presented and discussed, with particular focus on wall facing lateral displacements and bridge seat settlements. The effect of bridge load on the deformation behavior of the GRS bridge abutment was also investigated. Numerical results indicate that the abutment has good performance under static loading conditions with relatively small lateral deflection and settlement, and relatively large load bearing capacity.
1503
IFCEE 2015 © ASCE 2015
INTRODUCTION Geosynthetic-Reinforced Soil (GRS) walls have been used extensively for highway infrastructure projects and have many advantages over traditional retaining walls, including cost saving, relatively easy and fast construction, and good performance under static and seismic loadings. In recent years, GRS walls have also been proposed as bridge abutments with bridge loads supported on shallow foundations. This has significant cost savings as compared to conventional pile-supported designs and can reduce differential settlement between the bridge abutment and approach roadway (Abu-Hejleh et al. 2002). The GRS bridge abutment is a complex system that includes a lower GRS wall, bridge seat, and upper GRS wall, with the bridge superstructure load applied directly to the backfill for the lower GRS wall. The Colorado Department of Transportation (CDOT) completed a pioneering project for the Founders/Meadows Bridge in 1999. In this project, GRS walls were used to support the bridge superstructures and approach roadways. Maximum outward displacement of the lower wall facing and settlement of the bridge footing due to the placement of bridge superstructures were 10 mm and 13 mm, respectively. Post-construction monitoring results also indicated very good in-service performance for these GRS bridge abutments (Abu-Hejleh et al. 2002). In general, deformations of the instrumented abutments were small both during construction and after opening to traffic. GRS bridge abutments are typically subjected to much larger loads from bridge superstructures than conventional GRS walls. Numerical modeling has been widely used to study these effects. Hatami and Bathurst (2006) investigated the response of GRS walls under surcharge loading using finite difference method. Helwany et al. (2003) and Fakharian and Attar (2007) verified their numerical models against measured results from the Founders/Meadows Bridge abutments. Wu et al. (2006a) studied the allowable bearing capacity of GRS bridge abutments with flexible facing using finite element analysis. Zheng et al. (2014) also studied the behavior of a GRS bridge abutment under working stress condition. This paper presents a numerical investigation of the deformation behavior for a typical GRS bridge abutment during construction and service. The results mainly focus on the lower wall facing lateral displacement and bridge seat settlement. The effect of bridge load on the deformation of GRS bridge abutment is also investigated. NUMERICAL MODEL The finite difference computer program FLAC-2D version 7.0 (Itasca, 2011) was used to simulate the static response of a GRS bridge abutment. The abutment consists of a lower GRS wall, bridge seat, bridge superstructure loads, and an upper GRS wall.
1504
IFCEE 2015 © ASCE 2015
1505
Model Configuration and Finite Difference Grid The geometry and finite difference grid for the GRS bridge abutment model are shown in Figure 1. The foundation soil has a depth of 5 m below the GRS bridge abutment system. The total GRS bridge abutment height is H = 7.4 m, including a 5 m-high lower GRS wall with modular block facing and a 2.4 m-high upper wrapped-face GRS wall. The dimensions of concrete modular block facing elements are 0.3 m (width) × 0.2 m (height). The vertical spacing of geogrid layers in the lower GRS wall is 0.2 m. This close spacing of 0.2 m is a default value in the design of GRS bridge abutments with flexible facing, and the maximum reinforcement spacing is limited to 0.4 m to ensure satisfactory performance (Wu et al. 2006b). Geogrid reinforcement is rigidly connected to the facing blocks. An L-shaped bridge seat rests on top of the reinforced zone and is offset 0.4 m from the back of the lower wall facing blocks. Both concrete sections of the L-shaped bridge seat are 0.4 m thick and the loaded area has a width of 1.6 m. A 2.4 m-high wrapped-face GRS wall with a vertical spacing of 0.4 m is behind the bridge seat. The length of geogrid reinforcement is 5.18 m for both the lower and upper GRS walls, which is equal to 0.7 times the total bridge abutment height H .
2m
29.6 m Traffic Surcharge
Load 2.4 m 4
5m 5m
2 1
5m
5.18 m
3
Fig. 1. Model Configuration and Finite Difference Grid for GRS Bridge Abutment (1: Soil-Block Interface, 2: Block-Block Interface, 3: Soil-Geogrid Interface, 4: Soil-Bridge Seat Interface)
IFCEE 2015 © ASCE 2015
1506
Boundary Conditions The rear boundary for the FLAC model is located at a distance of 29.6 m (4 H ) behind the lower wall facing to minimize the influence of boundary conditions on the results. The front and rear boundaries of the model are fixed in the horizontal direction and free to move in the vertical direction, whereas the bottom boundary is fixed in both horizontal and vertical directions. Material Models and Properties The concrete facing blocks and bridge seat were modeled as elastic materials with elastic modulus E = 20 GPa and Poisson’s ratio ν = 0.2. A very stiff foundation soil, representing bedrock, was used in the simulations. Since the bedrock is very stiff, the foundation soil has little influence on the behavior of GRS bridge abutment. The backfill soil is specified as CDOT Class 1 structural backfill for the Founders/Meadows GRS bridge abutment, and consists of a mixture of gravel (35 %), sand (54.4 %), and fine-grained soil (10.6 %) (Abu-Hejleh et al. 2000). The backfill soil was modeled as an elastoplastic dilatant material with Mohr-Coulomb failure criterion. The Duncan-Chang hyperbolic relationship (Duncan et al. 1980) was implemented to simulate the non-linear stress-strain behavior of backfill soil prior to failure. Parameters for the backfill soil were calibrated with measured results from large-size triaxial tests (Abu-Hejleh et al. 2000). A summary of parameters for the backfill soil is presented in Table 1. Geogrid reinforcement was modeled using cable elements with tensile stiffness J = 1000 kN/m and yield strength T f = 100 kN/m. The soil-geogrid interfaces were characterized using c = 0 and φ = 39.5°, such that interface sliding was possible. The soil-block, block-block, and soil-bridge seat interfaces were modeled using interface elements. The friction angle for interfaces between soil and concrete (soil-block and soil-bridge seat interfaces) was assumed to be 2/3 φ (26°). The block-block interface friction angle was 35°, following results of direct shear tests reported by Ling et al. (2010). Table 1.
Model Parameters for Backfill Soil
γ (kN/m3)
K
n
Rf
Kb
m
pa (kPa)
c (kPa)
φ (°)
ψ (°)
22.1
1000
0.6
0.72
800
0
100
69.8
39.5
6
Modeling Procedures The construction sequence of the GRS bridge abutment was reproduced for the numerical simulations and performance was evaluated with static bridge loads and
IFCEE 2015 © ASCE 2015
traffic loads. The numerical simulations were performed in five stages: Stage 1: Construction of lower GRS wall with modular block facing. Stage 2: Placement of bridge seat. Stage 3: Construction of upper wrapped-face GRS wall. Stage 4: Placement of bridge loads. Stage 5: Application of traffic loads. The construction process for the lower GRS wall was simulated numerically by placing “lifts” with gravitational forces activated after each such placement. The thickness of each lift was equal to the height of one facing block (0.2 m), thus a total of 25 lifts were used to simulate construction of the lower wall. The soil-block and block-block interfaces were placed at specified positions and the geogrid layers were rigidly connected to the facing blocks at appropriate elevations. For field structures, the top three facing blocks are usually grouted together to maximize wall stability. A large tensile strength was assigned to the interfaces between these three blocks to numerically simulate the grouting effect. The L-shaped bridge seat was placed on top of the reinforced soil zone and the wrapped-face GRS wall was also constructed using successive lifts. In stage 4, a constant uniform vertical stress of 200 kPa was applied to the bridge seat over a loaded width of 1.6 m to simulate the vertical bridge load. This value is equal to the maximum allowable bearing pressure specified by FHWA design guidelines for GRS bridge abutments (Berg et al. 2009). A uniform vertical surcharge stress of 20 kPa was applied on the top of the model to simulate traffic loads. SIMULATION RESULTS Numerical results for GRS bridge abutment deformation, including wall facing lateral displacement and bridge seat settlement, after construction of lower GRS wall (Stage 1), placement of bridge loads (Stage 4), and opening to traffic (Stage 5), are presented in the following sections. Wall Facing Lateral Displacement The lateral displacement profiles for the lower GRS wall facing at different stages are shown in Figure 2. Outward displacements were small after construction of the lower GRS wall, with a maximum value of 8.2 mm near mid-height. After placement of bridge loads (Stages 4), the maximum lateral displacement increased to 17.8 mm at elevation 3.6 m. This trend is consistent with field monitoring results for the Founders/Meadows GRS bridge abutment, which showed maximum lateral displacement at the upper third of the lower wall after bridge construction (Abu-Hejleh et al. 2002). The maximum lateral displacement resulting from placement of the bridge loads (13.1 mm) occurred at the top of the wall. After
1507
IFCEE 2015 © ASCE 2015
1508
opening to traffic, wall facing lateral displacement increased approximately 3 mm for the upper third of wall. The maximum lateral displacements and corresponding elevations are summarized in Table 2. The location of maximum lateral displacement moved upward during construction, which indicated that the bridge superstructures induced more stress at the top portion of the wall, as the bridge load was applied directly on top of the reinforced soil mass. 5
Elevation (m)
4
3
2 Stage 1 Stage 4 Stage 5
1
0
Fig. 2.
Table 2.
0
5 10 15 20 25 Facing Lateral Displacement (mm)
30
Lateral Displacements of Lower Wall Facing
Maximum Wall Facing Lateral Displacement at Different Stages
Stage Maximum Lateral Displacement of Lower Wall Facing (mm) Elevation of Maximum Lateral Displacement (m)
1
2
3
4
5
8.2
9.0
13.8
17.8
21.1
2.4
2.6
3.0
3.6
3.6
Bridge Seat Movement Settlements and rotations of the bridge seat at different stages are summarized in Table 3. Each settlement represents an average of values at the toe and heel of the bridge seat structure after the construction of the lower wall. The bridge seat had 4.5 mm of settlement before placement of the bridge loads. After the 200 kPa vertical stress was applied, settlement increased to 15.3 mm, which corresponds to 0.21 % of the total bridge abutment height. Additional settlement induced by traffic
IFCEE 2015 © ASCE 2015
1509
load was approximately 3 mm. In general, settlements for both end-of-construction and in-service conditions were small. The vertical bridge load also induced rotation of the bridge seat. Rotation at the end of bridge abutment construction was 0.2° counterclockwise. Vertical soil stress under bridge seat after construction and application of traffic load are shown in Figure 3. After bridge abutment construction (Stage 4), vertical stress on the soil near the toe of the bridge seat reached a maximum of 220 kPa, which slightly exceeded the applied vertical stress (200 kPa). Stress then decreased approximately linearly to a value of 140 kPa at the heel. Non-uniform stress on the soil from the bridge seat resulted from asymmetry of the applied load about the centerline and lateral earth pressures from upper GRS wall on the bridge seat structure. Also, the average vertical stress on the soil is less than 200 kPa because the loaded width of the bridge seat (1.6 m) is less than the width of the bottom surface (2.0 m). The vertical stresses increased approximately 20 kPa after application of traffic load surcharge. Tab. 3. Settlement and Rotation of Bridge Seat at Different Stages Stage Bridge Seat Settlement (mm) Bridge Seat Rotation (°)
2
3
4
5
1.7
4.5
15.3
18.2
0.02
0.05
0.20
0.23
300
Vertical Stress on Soil (kPa)
Stage 4 Stage 5
250 200 150 100 50 0 0
2
4 6 8 Distance from Facing (m)
10
Fig. 3. Vertical Stresses under Bridge Seat
IFCEE 2015 © ASCE 2015
1510
EFFECT OF BRIDGE LOAD Wu et al. (2006a) studied the allowable bearing pressure for GRS bridge abutments using finite element analysis. They performed a series of parametric studies and found that abutments with a reinforcement spacing of 0.2 m could sustain vertical stresses up to 1000 kPa without catastrophic failure. The effect of bridge load on GRS bridge abutment deformation was also investigated using FLAC. Maximum facing lateral displacements and bridge seat settlements are shown in Figure 4 for vertical stresses up to 1000 kPa. The maximum lateral displacement and bridge seat settlement before application of the bridge load are 13.8 mm and 4.5 mm, respectively. Figure 4(a) shows an approximate linear relationship between maximum lateral displacement and bridge load for applied stress levels up to 1000 kPa, where the maximum displacement was 44 mm. Figure 4(b) shows a similar linear increase for bridge seat settlement. The settlement reached 62 mm under 1000 kPa, which corresponds to 0.84 % of the total bridge abutment height. 80 70
40 60 Settlement (mm)
Maximum Lateral Displacement (mm)
50
30
20
50 40 30 20
10 10
(a) 0
(b)
0
0
200
400 600 Bridge Load (kPa)
800
1000
0
200
400 600 Bridge Load (kPa)
800
1000
Fig. 4. Effect of Bridge Load on (a) Facing Maximum Lateral Displacement; (b) Bridge Seat Settlement CONCLUSIONS The deformation behavior of a GRS bridge abutment during construction and service was numerically simulated using the finite difference program FLAC-2D. The GRS abutment was subjected to a vertical stress of 200 kPa to simulate the load from bridge superstructures to the top of reinforced soil zone. Numerical simulations indicated that the GRS abutment had a maximum lateral displacement of 17.8 mm for the lower wall facing and an average settlement of 15.3 mm for the bridge seat. Additional settlement of 3 mm for bridge seat occurred due to
IFCEE 2015 © ASCE 2015
application of traffic loads. Overall, the numerical simulations indicate very good performance at the end of construction and under service load conditions. The results also indicate that this GRS bridge abutment, with closely spacing reinforcement, could sustain vertical stresses up to 1000 kPa with relatively small deformations (44 mm lateral, 62 mm vertical). ACKNOWLEDGEMENTS Financial support for this investigation was provided by the California Department of Transportation (Caltrans) and is gratefully acknowledged. We also thank Dr. Charles S. Sikorsky of the Caltrans Office of Earthquake Engineering for his support and assistance with the project. REFERENCES Abu-Hejleh, N., Outcalt, W., Wang, T., and Zornberg, J.G. (2000). “Performance of Geosynthetic-Reinforced Walls Supporting the Founders/Meadows Bridge and Approaching Roadway Structures, Report 1: Design, Materials, Construction, Instrumentation, and Preliminary Results.” Report No. CDOT-DTD-R-2000-5, Colorado Department of Transportation, Colorado, USA. Abu-Hejleh, N., Zornberg, J.G., Wang, T., and Watcharamonthein, J. (2002). “Monitored displacements of unique geosynthetic-reinforced soil bridge abutments.” Geosynthetics International, 9 (1): 71-95. Berg, R.R., Christopher, B.R., and Samtani, N. (2009). “Design of mechanically stablized earth walls and reinforced soil slopes.” FHWA-NHI-10-024, Federal Highway Adminstration, Washington, D.C., USA. Duncan, J.M., Byrne, P., Wong, K.S., and Mabry, P. (1980). “Strength, stress-strain and bulk modulus parameters for finite element analysis of stresses and movements in soil masses.” Report No. UCB/GT/80-01, University of California, Berkeley, CA, USA. Fakharian, K., and Attar, I.H. (2007). “Static and seismic numerical modeling of geosynthetic-reinforced soil segmental bridge abutments.” Geosynthetics International, 14(4), 228-243. Hatami, K., and Bathurst, R.J. (2006). “Numerical model for reinforced soil segmental walls under surcharge loading.” Journal of Geotechnical and Geoenvironmental Engineering, 132(6), 673-684. Helwany, S.M.B., Wu, J.T.H., and Froessl, B. (2003). “GRS bridge abutments – an effective means to alleviate bridge approach settlement.” Geotextiles and Geomembranes, 21, 177-196. Itasca Consulting Group, Inc. (2011). “Fast Lagrangian Analysis of Continua - 2D Version 7.0.” Itasca Consulting Group, Inc., Minneapolis.
1511
IFCEE 2015 © ASCE 2015
Ling, H.I., Yang, S., Leshchinsky, D., Liu, H., and Burke, C. (2010). “Finite-element simulations of full-scale modular-block reinforced soil retaining walls under earthquake loading.” Journal of Engineering Mechanics, 136(5), 653-661. Wu, J.T.H., Lee, K.Z.Z., and Pham, T. (2006a). “Allowable bearing pressures of bridge sills on GRS abutments with flexible facing.” Journal of Geotechnical and Geoenvironmental Engineering, 132(7): 830-841. Wu, J.T.H., Lee, K.Z.Z., Helwany, S.B., and Ketchart, K. (2006b). “Design and construction guidelines for geosynthetic-reinforced soil bridge abutments with a flexible facing.” NCHRP Report 556, Transportation Research Board, Washington, D.C., USA. Zheng, Y., Fox, P.J., and Shing, P.B. (2014). “Numerical Simulations for Response of MSE Wall-Supported Bridge Abutment to Vertical Load.” GeoShanghai 2014, International Conference on Geotechnical Engineering 2014, Shanghai, China.
1512
