Seismic performance of bar-mat reinforced-soil retaining wall: Shaking table testing versus numerical analysis with modified kinematic hardening constitutive model

https://doi.org/10.1016/j.soildyn.2010.04.020Get rights and content

Abstract

Reinforced-soil retaining structures possess inherent flexibility, and are believed to be insensitive to earthquake shaking. In fact, several such structures have successfully survived destructive earthquakes (Northridge 1994, Kobe 1995, Kocaeli 1999, and Chi-Chi 1999). This paper investigates experimentally and theoretically the seismic performance of a typical bar-mat retaining wall. First, a series of reduced-scale shaking table tests are conducted, using a variety of seismic excitations (real records and artificial multi-cycle motions). Then, the problem is analyzed numerically employing the finite element method. A modified kinematic hardening constitutive model is developed and encoded in ABAQUS through a user-defined subroutine. After calibrating the model parameters through laboratory element testing, the retaining walls are analyzed at model scale, assuming model parameters appropriate for very small confining pressures. After validating the numerical analysis through comparisons with shaking table test results, the problem is re-analyzed at prototype scale assuming model parameters for standard confining pressures. The results of shaking table testing are thus indirectly “converted” (extrapolated) to real scale. It is shown that: (a) for medium intensity motions (typical of Ms≈6 earthquakes) the response is “quasi-elastic”, and the permanent lateral displacement in reality could not exceed a few centimeters; (b) for larger intensity motions (typical of Ms≈6.5–7 earthquakes) bearing the effects of forward rupture directivity or having a large number of strong motion cycles, plastic deformation accumulates and the permanent displacement is of the order of 10–15 cm (at prototype scale); and (c) a large number of strong motion cycles (N>30) of unrealistically large amplitude (A=1.0 g) is required to activate a failure wedge behind the region of reinforced soil. Overall, the performance of the bar-mat reinforced-soil walls investigated in this paper is totally acceptable for realistic levels of seismic excitation.

Introduction

Invented by the French Architect and Engineer Henri Vidal in the late 50s, “reinforced earth” can be characterized as a composite material. It combines the compressive and shear strength of a thoroughly compacted “select” granular fill (with specific requirements concerning grain distribution, fines content, plasticity index, friction angle, etc.) with the tensile strength of reinforcing materials, such as mild steel (e.g. dip galvanized flat ribbed strips or welded wire mats) or geosynthetic polymers (polypropylene, polyethylene, or polyester geogrids, or woven and non-woven geo-textiles). The latter compensates for the weak strength of soil in tension, rendering reinforced earth the direct analog of reinforced concrete in soil. Depending on the nature of the reinforcement, a reinforced earth system may be characterized as inextensible (when the reinforcement fails without stretching as much as the soil) or extensible (when the opposite is true). Inextensible steel reinforcements are most common for critical structures, such as bridge abutments where control of deformation is crucial. On the other hand, extensible geosynthetic reinforcement is often used in reinforced slopes, basal reinforcement, and temporary retaining walls, where there is no concern for displacement.

Reinforced earth retaining walls posses a number of technical and economic advantages compared to standard gravity walls: (a) they can be constructed rapidly, without requiring large construction equipment; (b) they require less site preparation and less space in front of the structure for construction operations, thus reducing the cost of right-of-way acquisition; (c) they do not need rigid foundation support as they are tolerant to deformations; and (d) they are very cost effective and technically feasible even for heights exceeding 25 m. The first such wall in a seismically active area was constructed in California’s Sate Highway 39, in 1972. Since then, in recognition of all the previously discussed advantages, their use quickly spread universally in highway, industrial, military, commercial, and residential applications.

Reinforced earth structures have all the necessary “ingredients” to be earthquake resistant: being flexible, they tend to follow the dynamic deformation of the retained (free-field) soil without attracting substantially large dynamic earth pressures (e.g. [1]). Indeed, several reinforced soil walls have experienced large intensity destructive earthquakes (Loma Prieta 1989, Northridge 1994, Kobe 1995, Chi-Chi 1999, and Kocaeli 1999) without considerable damage. One of the most dramatic such examples is the 1994 Mw 6.8 Northridge earthquake. With many recorded PGA values higher than 0.60 g, the inflicted damage to structures of all kinds was rather extensive, while 5 major freeway bridges, 18 parking stations, and 40 buildings totally collapsed. Surprisingly, the damage to 23 reinforced soil walls of several heights within the affected area of the earthquake was minor [2]. Regardless of their location and recorded level of PGA, all of them were found to be fully intact with no conspicuous structural damage. Only in one case, minor concrete spalling on the facing panel was observed.

Even more interesting is the performance of reinforced earth walls during the 1995 Mw 7 Kobe earthquake. With recorded PGAs exceeding 0.8 g, the damage was devastating with the direct economic loss exceeding $100 billion [3], [4], [5], [6]. The damage to all sorts of structures was more than devastating: from the Kobe Port which was practically put out of service (all but 7 of its 186 berths were totally damaged) to the spectacular overturning structural collapse of a 630 m section the elevated Hanshin Expressway, to countless collapses of bridges and buildings, and to numerous landsides. Also substantial was the damage to a variety of gravity-type retaining structures [7], [8], [9], [10]. In marked contrast, damage to reinforced earth walls was rather minor [11], [12]. A total of 124 reinforced earth structures, of height ranging from 2 to 17 m were inspected after the earthquake. Although most of them had been designed for ground acceleration of the order of 0.15 g, 74% of them sustained no damage at all, 24% had only very minor damage (mainly displacement), and only 2% showed some damage to the wall facing and movement of the retained soil. No collapse or clear failure was observed.

The seismic performance of reinforced earth structures has been investigated experimentally with various methods: from soil element testing [13], to centrifuge model testing [14], [15], [16], [17], [18], [19], [20], [21], and shaking table testing at reduced [22], [23], and at nearly full scale [18], [24], [25], [26]. Among the several conclusions

  • (i)

    the critical acceleration is a function of backfill density [21];

  • (ii)

    the stiffness, spacing, and length of the reinforcement directly affect the stability and the lateral and vertical deformation of the wall [15], [17], [18], [19], [21];

  • (iii)

    the length of the reinforcement is not crucial, as long as it exceeds 70% of the wall height [21];

  • (iv)

    the backfill is subjected to substantial densification and settlement [19], [21];

  • (v)

    current pseudo-static seismic stability analyses based on the limit equilibrium method underestimate their seismic stability [24], [27];

  • (vi)

    the largest lateral displacement takes place at the middle-height of the wall [19]; and

  • (vii)

    finite element (FE) simulation can capture the dynamic response of reinforced earth walls, provided that nonlinear soil response is modeled with a realistic constitutive law [18], [25].

This paper investigates experimentally and analytically the seismic response of typical reinforced soil (bar-mat) retaining walls. First, we present the experimental setup and the key results of a series of reduced-scale shaking table testing. Then, a nonlinear FE model is developed for the same problem. A modified kinematic hardening model is developed and encoded in ABAQUS through a user subroutine. The parameters are calibrated through experimental data (soil element testing of the “Longstone” sand used in the experiments): (a) for small confining pressures (which are considered representative for the 1g shaking table tests), and (b) for standard confining pressures (which are considered representative for the prototype problem). First, we analyze the shaking table test (assuming model parameters for small confining pressures) to validate the analysis methodology and the constitutive model. Then, we analyze the prototype (assuming model parameters for standard confining pressures), thus extending our results to the real scale.

Section snippets

Shaking table testing

A series of two models were constructed and tested at the Laboratory of Soil Mechanics of the National Technical University of Athens (NTUA), utilizing a recently installed shaking table. The table, 1.3 m×1.3 m in dimensions, is capable of shaking specimens of 2 tons at accelerations upto 1.6 g. Synthetic accelerograms, as well as real earthquake records can be simulated. The actuator is equipped with a servo-valve, controlled by an analog inner-loop control system and a digital outer-loop

Results of shaking table testing

In the following sections we present the performance of the tested bar-mat reinforced soil walls under extreme seismic shaking and under more realistic seismic motions. As already discussed, an additional substantial difference between the two test series lies in the relative density of the retained soil: loose and dense sand, respectively.

Numerical analysis

Two sets of numerical analysis are conducted: (i) analysis of the shaking table model, assuming soil parameters measured for small confining pressures; and (ii) analysis of the prototype, assuming realistic soil parameters for standard confining pressures. The first set of analysis is aimed to corroborate the numerical method and the modified kinematic hardening constitutive soil model. Then, the validated numerical methodology is utilized to predict the actual performance of the prototype.

Results of numerical analysis and interpretation

The results of the numerical analyses are summarized in Fig. 18, Fig. 19, and Table 3. The results are shown for both sets of analysis (analysis of the shaking table test, assuming model parameters for small confining pressures; and analysis of the prototype, assuming model parameters for standard confining pressures). The first set (analysis of the shaking table model) is compared directly with shaking table test results, to serve as validation of the numerical analysis and of the modified

Summary and conclusions

This paper has investigated experimentally and numerically the seismic performance of typical bar-mat-reinforced soil retaining walls. The main conclusions of the work presented herein are as follows:

  • [1]

    Although the stress field in the backfill soil cannot be correctly reproduced in reduced-scale shaking table testing, the latter can be used to simulate the behavior of reinforced soil walls, provided that the results are interpreted carefully, with due consideration to scale effects and the stress

Acknowledgements

This work forms part of an EU 7th Framework research project funded through the European Research Council’s Programme “Ideas”, Support for Frontier Research—Advanced Grant under Contract number ERC-2008-AdG228254-DARE to Professor G. Gazetas. The authors gratefully acknowledge the encouragement and guidance of Professor Gazetas throughout this research effort.

References (45)

  • J. Koseki

    Seismic performance of retaining walls-case histories and model tests

    Proceedings of the fourth forum on implications of recent earthquakes on seismic risk

    (2002)
  • Kobayashi K, Tabata H, Boyd M. The performance of reinforced earth structures in the vicinity of Kobe during the Great...
  • F Tatsuoka et al.

    Seismic stability against high seismic loads of geosynthetic reinforced soil retaining structures

    (1998)
  • R.L. Michalowski et al.

    Triaxial compression of sand reinforced with fibers

    Journal of the Geotechnical Engineering Division, ASCE

    (2003)
  • J.G. Zornberg et al.

    Performance of geosynthetic reinforced slopes at failure

    Journal of the Geotechnical Engineering Division

    (1998)
  • A Takahashi et al.

    Dynamic behavior of vertical geogrid-reinforced soil during earthquake

    (1999)
  • J.G. Zornberg et al.

    Strain distribution within geosynthetic-reinforced slopes

    Journal of Geotechnical and Geoenvironmental Engineering, ASCE

    (2003)
  • S. Ichikawa et al.

    Centrifuge model tests on seismic stability of reinforced retaining wall

    The fifth workshop on safety & stability of infrastructures against environmental impacts

    (2005)
  • H.I. Ling et al.

    Analyzing dynamic behavior of geosynthetic-reinforced soil retaining walls

    Journal of Engineering Mechanics, ASCE

    (2004)
  • R.V. Siddharthan et al.

    Seismic deformation of bar mat mechanically stabilized earth walls—centrifuge tests

    Journal of Geotechnical and Geoenvironmental Engineering

    (2004)
  • Y Ishihama et al.

    Centrifuge tilting and shaking table tests on the RSW with different soils

    Proceedings of the third Asian regional conference on geosynthetics (GeoAsia 2004)

    (2004)
  • L. Nova-Roessig et al.

    Centrifuge model studies of the seismic response of reinforced soil slopes

    Journal of Geotechnical and Geoenvironmental Engineering, ASCE

    (2006)
  • Cited by (0)

    View full text