Search for: All records

This paper investigates and presents the numerical modeling and validation of the response of a uniform clean sand using monotonic and cyclic laboratory tests as well as a centrifuge model test comprised of a submerged slope. The dynamic response of the sand is modeled using a critical state compatible, stress ratio-based, bounding surface plasticity constitutive model (PM4Sand), implemented in the commercial finite-difference platform FLAC, and PM4Sand’s performance is evaluated against a comprehensive testing program comprised of laboratory data and a well-instrumented centrifuge model test. Three different calibrations informed by the lab and centrifuge data are performed and the goodness of the predictions is discussed. Conclusions are drawn with regards to the performance of the simulations against the laboratory and centrifuge data, and recommendations about the calibration of the model are provided. 

A cone penetrometer was specifically designed for the LEAP project to provide an assessment of centrifuge model densities independent from mass and volume measurements. This paper presents the design of the CPT and analyses of the results. Due to uncertainty in the specifications about how to define zero depth of penetration, about 20% of the CPT profiles were corrected to produce more accurate results. The procedure for depth correction is explained. After these corrections, penetration resistances at the representative depths of 1.5, 2, 2.5, and 3 m (prototype depth) are correlated to the reported specimen dry densities by linear regression. Using the inverse form of the linear regression equations, the density of each specimen could be estimated from the penetration resistance. Kutter et al. (LEAP-UCD-2017 comparison of centrifuge test results. In Model Tests and Numerical Simulations of Liquefaction and Lateral Spreading: LEAP-UCD-2017, 2019b) found that the density determined from penetration resistance was a more reliable predictor of liquefaction behavior than the reported density itself. Finally, the centrifuge tests at different LEAP facilities modeled the same prototype in different containers using different length scale factors (1/20 to 1/44); thus the ratio of layer thickness to cone diameter was different in each experiment. It appears that the penetration resistances are noticeably affected by container width and, to a lesser extent, resistance is affected by the length scale factor. 

Three centrifuge experiments were performed at the University of California, Davis, for LEAP-UCD-2017. LEAP is a collaborative effort to assess repeatability of centrifuge test results and to provide data for the validation of numerical models used to predict the effects of liquefaction. The model configuration used the same geometry as the LEAP-GWU-2015 exercise: a submerged slope of Ottawa F-65 sand inclined at 5 degrees in a rigid container. This paper focuses on presenting results from the two destructive ground motions from each of the three centrifuge models. The effect of each destructive ground motion is evaluated by accelerometer recordings, pore pressure response, and lateral deformation of the soil surface. New techniques were implemented for measuring liquefaction-induced lateral deformations using five GoPro cameras and GEO-PIV software. The methods for measuring the achieved density of the as-built model are also discussed. 

This paper presents numerical liquefaction simulations of a hypothetical sand in cyclic direct simple shear tests for four different constitutive models. The four models are PM4Sand in FLAC, UBCSand in FLAC, Pressure Dependent Multi Yield 02 (PDMY02) in OpenSees, and the Manzari-Dafalias04 model in OpenSees. Parameters published by others for various sands were used for this work to avoid possible introduction of our bias into the calibration process. The material properties were determined by others for different uniformly graded sands, all with a relative density of approximately 50%. The simulations do not pertain to one sand or one set of laboratory data, so therefore, there is no right or wrong answer. Instead, the goal of this paper is compare a consistent set of results that show the implementations of each model behave as expected, and to illustrate basic differences in behavior of the different models. Cyclic strength curves (cyclic stress as a function of number of cycles) illustrate the behavior of the models over a range of cyclic stresses. Each model displays pore pressure build up, softening, and cyclic contraction-dilation cycles associated with cyclic mobility. Two of the models soften to a point, but then stabilize in a repeated hysteresis loop with no additional growth in the cyclic strain amplitude after some number of cycles. 

Twenty-four centrifuge model tests have been conducted at nine different geotechnical centrifuge facilities around the world as part of the international LEAP effort (liquefaction experiments and analysis projects). All of the centrifuge models represent a 4 m deep 5 degree sloping submerged sand deposit. The mean effective PGA of the input motion for all of the experiments was approximately 0.15 g and the mean relative density was approximately 65%, but the effective PGA’s varied between about 0.07 g and 0.3 g, and the relative densities varied between about 40% and 75%. The test matrix was designed to enable experimental quantification of not only the median response but also the trend and sensitivity of the model response to density and shaking intensity. Quantification of the sensitivity of the response to initial conditions is a prerequisite for objective evaluation of the quality of the model test data. In other words, a discrepancy between two experiments should be evaluated after accounting for the uncertainty in the initial conditions and the sensitivity of the response to initial conditions. For the first time, a sufficient number of experiments has been performed on a similar problem to provide meaningful quantitative evaluation of the trend between PGA, density, and displacement. The sensitivity is quantified by the gradient of the trend and the uncertainty of the trend is quantified from the residuals between the fitting data and the trend.