Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 73
PROCEEDINGS OF THE EIGHTH INTERNATIONAL CONFERENCE ON CIVIL AND STRUCTURAL ENGINEERING COMPUTING
Edited by: B.H.V. Topping
Paper 109

Finite Element Predictions of Centrifuge Tests on Liquefiable Reinforced Soils

O.O.R. Famiyesin, A.A. Rodger and A. Matheson

Department of Engineering, University of Aberdeen, United Kingdom

Full Bibliographic Reference for this paper
O.O.R. Famiyesin, A.A. Rodger, A. Matheson, "Finite Element Predictions of Centrifuge Tests on Liquefiable Reinforced Soils", in B.H.V. Topping, (Editor), "Proceedings of the Eighth International Conference on Civil and Structural Engineering Computing", Civil-Comp Press, Stirlingshire, UK, Paper 109, 2001. doi:10.4203/ccp.73.109
Keywords: reinforced soil, finite element, liquefaction, earthquake, centrifuge, geotextiles, geogrids.

Summary
This paper shows how a reinforced saturated soil system would behave in the event of an earthquake, by the finite element (FE) analysis of modified forms of two previous centrifuge model tests. A horizontal and a vertical layer of reinforcement were respectively embedded within the soil medium in the two centrifuge models. The effect of the reinforcement properties on the modified system was studied by monitoring the surface deformation, excess pore pressure variation and the fundamental frequency of acceleration response. Consistency of the predictions was assured by using different levels of mesh refinement and mesh patterns.

The two centrifuge tests modified for this FE study, were respectively named KVV03, (by Venter [1] at the Cambridge Geotechnical Centrifuge UK), and VELACS (Model No 3, carried out at California Institute of Technology Pasadena [2], University of California Davis [3] and Rensselaer Polythecnic Institute Troy [4]). The KVV03 model [1] consisted of a concrete dyke lying on a silicon-oil, fully saturated sand layer built into a strong box, with an oil reservoir created behind the dyke to provide seepage conditions. The VELACS (No 3) model [2,3,4] consisted of two vertical, water-saturated sand layers (of relative densities Dr = 40% and 70% respectively), contained in a laminar box. Both tests were carried out carried out under 78g and 50g conditions respectively with their earthquake traces applied as base motion. The modified KVV03 and VELACS models were used respectively for the horizontal and vertical reinforcement study.

Chan, Famiyesin and Muir Wood have carried out successfully FE simulation of the KVV03 test [5] and a Class A prediction of the VELACS (No 3) test [6] using DIANA-SWANDYNE II [7], a Fortran 77 program based on the u-p approximation of the Biot formulation. This program has access to the Pastor-Zienkiewicz mark-III soil model which is suitable for soils under earthquake loading conditions. This FE program and soil model were used for all the analyses carried out in this study. An initial stress static analysis and a no-earthquake dynamic equilibrium analysis were carried out to ensure that the insitu stresses within the soil were in equilibrium with the applied gravity load. The earthquake trace was applied in the dynamic analysis stage while the post-earthquake, consolidation analysis leads to pore pressure dissipation. The modified centrifuge models were analysed and the influence of the reinforcement properties on the soil behaviour studied both during the dynamic and consolidation phases [8].

The results show that the presence of reinforcement and its increase in stiffness do significantly reduce the vibration of the structure on the soil surface and also reduces its immediate and longer term settlements. A higher reinforcement `permeability' was noted to result in a reduction of both the time required for complete pore pressure dissipation and the settlement of the supported structure. Reinforcement 'permeability' values of the same order as that of the surrounding soil did not produce any significant changes within the soil or in the supported structure. Variations in other reinforcement parameters such as the yield stress, Poisson ratio and hardening parameter did not significantly affect the soil response under earthquake conditions. The outcomes of this study may be of interest to design engineers and geotextile manufacturers who may be looking for new ways of using the material within the construction industry, especially in earthquake prone regions of the world.

References
1
K.V. Venter, "Modelling the response of sand to cyclic loads", PhD Thesis, Cambridge University, 1987.
2
R.F. Scott, B. Hushmand and H. Rashidi, "Model No 3 primary test description and test results", In Verification of numerical procedures for the analysis of soil liquefaction problems, Eds. K.Arulanandan & R.F.Scott, Balkema Rotterdam, 1, 435-462, 1993.
3
T.M. Farrell and B.L. Kutter, "Experimental results for Model No 3", ibid, 1, 463-469, 1993.
4
V.M. Taboada and R. Dobry, "Experimental results of attempted duplication of Model No 3 at RPI", ibid, 1, 471-482, 1993.
5
A.H.C Chan, O.O. Famiyesin and D. Muir Wood, "Detailed comparison of the explicit u-w and implicit u-p formulations for the dynamic analysis of saturated soils", Report No. CE-GE93-32, Dept. of Civil Eng., Glasgow University, 1993.
6
A.H.C Chan, O.O. Famiyesin and D. Muir Wood, "Numerical simulation report for the VELACS project", Report No. CE-GE92-23, Dept. of Civil Eng., Glasgow University, 1992.
7
A.H.C. Chan, "User manual for DIANA-SWANDYNE II", Department of Civil Eng., University of Glasgow, 1990.
8
A. Matheson, "Finite element analysis of reinforced soil under earthquake loading", BEng Thesis, University of Aberdeen, 1997.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £122 +P&P)