Research Experience

• Investigation of seismic integrity of rammed aggregate piers at sites prone to large deformations using OpenSees

Rammed Aggregate Piers (RAPs) are a proprietary ground improvement system (Geopier®) in which compacted aggregates are used to create stiff pier elements. The objective of this research was to investigate the behavior of RAP elements under seismic loading. As a result of ground shaking, the soil and consequently the RAP elements will be subjected to shear straining. This study focused on obtaining a first order approximation on the magnitude of stresses in RAP elements when they interact with the surrounding soil when subjected to ground shaking. For the analyses, equivalent linear site response analyses (using Strata) were performed to calculate displacement time histories in the soil. These displacement time histories then were applied to the RAP element to calculate moments and stresses in the RAP element using OpenSees. The RAP element was discretized in nonlinear beam-column elements. For the bending moment, the non-linear constitutive model of Bouc-Wen-Gerolymos-Gazetas (BWGG uniaxial material) and elastic uniaxial material was considered. For the simulation of the soil, truss elements with spring and dashpot were considered. The dashpot was assumed to be a linear viscous material while the spring was a BWGG uniaxial material. Numerical results showed that bending moments in RAP element, in most of analyses, exceeded the moment capacity. Numerical results also indicated that the computed bending moments in the RAP element are sensitive to top moment fixity of RAP element, shear wave velocity profile, and input bedrock motion.

• Treatment of Epistemic Uncertainty in Site Effects in Probabilistic Seismic Hazard Analysis

The focus of this project was on the treatment of epistemic uncertainty, primarily in shear wave velocity profile, associated with site effects in probabilistic seismic hazard analysis (PSHA). One of the guiding rules in estimating epistemic uncertainty is that the less data is available, the larger the estimated uncertainty should be. The implication of this rule is that higher uncertainty will lead to a higher computed seismic hazard. However, contrary to the intrinsic implication of this guiding rule, the treatment of epistemic uncertainty for site effects in PSHA in the current methodology provided in EPRI (2012) can result in a reduction of calculated ground motions as epistemic uncertainty increases. Per EPRI (2012), epistemic uncertainty in the profile stratigraphy is taken into account by developing best estimate, lower bound, and upper bound base cases for the shear wave velocity profiles for the site. The base case Vs profile are used in the site response analysis to obtain site amplification factors (AF) for each base case and a weighted average AF curve is computed from the base cases. Comparing the average AF curves calculated from the base profiles per EPRI (2012) for two different levels of uncertainty indicates that the bandwidth of the weighted average amplification curves is wider than those of the base cases profiles, and the amplitude of the weighted average curve for lower level of uncertainty is greater than that for the higher level of uncertainty for a range of periods. In this project, an alternative procedure was proposed to normalize amplification factor curves by fundamental period of the profile (Tp) resulting in the calculated weighted average amplification curves having a bandwidth and amplitude that are similar to the curves for bases profiles.

• Literature review on shear strength of low-plasticity silts

In this project, a literature review on the main geotechnical aspects of silts was performed. Classification and composition, consolidation behavior, stress-strain, and critical state behavior of different silts from different places including US, Italy, India, and Turkey were summarized. In addition, different failure criteria that can be used for silts, and the shear strength measured from these failure criteria were also presented. Effects of cavitation especially on dilatant silts and the main potential causes of cavitation in undrained tests was also reviewed.

• Analysis of active earth pressure on retaining walls due to strip foundations

The main purpose of this project was to provide a rigorous solution to the problem of active earth pressure in the framework of upper-bound theorem of limit analysis to produce design charts for calculation of the lateral active earth pressure of backfill when loaded by a strip foundation. A kinematically admissible collapse mechanism consisting of several rigid blocks with translational movement was considered in which energy dissipation take place along planar velocity discontinuities. These results were presented in the form of dimensionless design charts relating the mechanical characteristics of the soil, strip load conditions, and the active earth pressure.

• Three dimensional analysis of active earth pressure on retaining walls induced by a locally loaded backfill

This research pertains to a solution based on the upper-bound theorem of limit analysis to calculate 3D active earth pressure induced by a homogeneous frictional-cohesive locally-loaded backfill. A failure mechanism consisting of rigid blocks that move with constant velocities was assumed as the kinematically admissible velocity field. Energy dissipation was restricted to areas at the interface of adjacent blocks, and interface of blocks in motion and soil at rest. The numerical results of 3D active earth pressure were presented in the form of dimensionless charts for various governing parameters for practical use in geotechnical engineering.

The end effects on active earth pressure problems was also evaluated. In some cases, where the failure mechanism extends to a reasonably long distance, conventional plane-strain analysis can be applied to observed three-dimensional failures. However, some problems in which the potential failure mechanism is constrained and, for locally loaded backfills and slopes, the end effects are of paramount importance, and three-dimensional analysis is necessary. Comparing 2D and 3D analyses showed that 2D analysis greatly overestimates the active earth pressure. Numerical results indicated that for backfills with higher internal friction angles or cohesions, and for foundations farther from the wall or with lower surcharges, the end effects are more pronounced.