LEAP-ASIA-2019 and LEAP-UCD-2017 centrifuge tests performed at the University Gustave Eiffel

Abstract

For the LEAP-UCD-2017 and ASIA-2019 exercises, an international effort has been made by centrifuge teams and modellers to validate modelling and simulation procedures for the triggering and the consequences of liquefaction phenomena. The configuration tested in the centrifuge was simple: a saturated sand slope with an inclination of 5° was subjected to one or to a series of 1Hz ramp sine motions of the same or different amplitudes. For each exercise, University Gustave Eiffel has performed 2 tests. In the framework of the objectives of both exercises, the two first tests performed, S02 and S03, enriched the database on the effect of density on the response of the slope and the two others, A02 and A03, provided data on the validity of the generalized scaling law (GSL) proposed by Iai et al. (2005). Furthermore, based on the density calculated from CPT correlation (Carey et al. 2020b), the comparison of the results of the tests S03 and A02 provide insight on the effect of the base shaking amplitude. After a description of the experimental setup and procedure, that includes an analysis of the obtained base shakings, an analysis of the boundary effects induced during the tests is first presented. The effects of the density, base shaking amplitude and the validity of GSL are analysed based on the pore pressure build up, the acceleration and the dynamic displacement obtained from a vertical array of horizontal accelerometers and pore pressure sensors located in the middle of the slope. In addition, their effects on the residual displacement at the slope surface and on the stress-strain loops are also presented. The analysis of the boundary effects highlights that they are limited for dense sand subjected to moderate base shaking but they become significant with the decrease of the relative density and/or the increase of the base shaking amplitude. In medium dense sand, the dissimilarities of the soil displacement near the extremities of the container and in the middle of the slope become significant and, for the same density, evidence of p-wave creation clearly appears with the increase of the base shaking amplitude. Despite evidence of boundary effects on accelerations, residual displacements at the surface and on strain-stress loops, the comparison of the tests highlights effects of density and base shaking on the pervasiveness of the liquefaction and the soil column behaviour. The GSL approach seems to be reliable concerning the qc value at 2 m depth and minor differences were observed on the time history of the accelerations and pore pressure in the middle of the slope. However, dissimilarities in the dynamic displacements and a difference of residual displacements at the soil surface were noticed. These results should be confirmed due to base shaking differences and noticeable boundary effects.