A Physics-Informed Semi-Empirical Probabilistic Model for Predicting Building Settlement on Liquefiable Ground
Shideh Dashti, Ph.D.
Geotechnical Engineering and Geomechanics
Civil, Environmental, and Architectural Engineering
University of Colorado Boulder
Abstract: This presentation introduces a predictive model for the settlement of shallow-founded structures on liquefiable ground during earthquakes. The model is based on the results of an extensive fully-coupled, 3D, nonlinear numerical parametric study of soil-foundation-structure systems, validated with centrifuge experiments as well as a database of case history observations. The state of practice still largely involves estimating building settlement using empirical procedures for volumetric strains in the free-field, which have been shown as unreliable and inappropriate. We performed a series of centrifuge experiments to evaluate the dominant mechanisms of deformation near shallow-founded structures and the influence of different testing parameters. Results from centrifuge tests were subsequently used to evaluate the predictive capabilities of 3D, fully-coupled, dynamic finite element analyses of the soil-foundation-structure system with the PDMY02 soil constitutive model implemented in OpenSees. A numerical parametric study (exceeding 63,000 simulations) followed to identify the most optimum Intensity Measures (IMs) for predicting permanent building settlement on liquefiable ground. The numerical study also provided insight into the relative importance and influence of different input parameter and the functional form of the predictive model for settlement. The collected case history database helped validate and refine the predictive model, accounting for complexities of the ground motion and site conditions in the field. Nonlinear regression and latent variable analysis were used to develop model coefficients. The uncertainty around model estimates was modeled by a Gumbel extreme value distribution. An additional logistic model was provided to estimate the probability of insignificant settlement (defined as less than 1 centimeter). The proposed procedure will be the first to consider variations in site conditions, building’s presence and properties, and the inherent uncertainties, which is a necessary step before the benefits of performance-based seismic design can be realized in the evaluation and mitigation of the liquefaction hazard.
Bio: Dr. Shideh Dashti is an Assistant Professor in Geotechnical Engineering and Geomechanics at the University of Colorado Boulder. She obtained her undergraduate degree at Cornell University and graduate degrees at the University of California, Berkeley. She worked briefly with ARUP (New York City) and Bechtel (San Francisco) Geotechnical groups on several engineering projects in the US and around the world involving the design of foundation systems, slopes, and underground structures and tunnels. Her research team at CU Boulder studies: the interactions and interdependencies among different infrastructure systems during earthquakes and other types of disasters; the seismic performance of underground structures; and consequences and mitigation of the liquefaction hazard facing structures in isolation and in dense urban settings.
Date(s) - Sep 01, 2017
2:00 pm - 3:00 pm