An analytical/empirical model for prediction of lateral spread displacements | | Posted on:2003-03-19 | Degree:Ph.D | Type:Dissertation | | University:University of Washington | Candidate:Baska, David Ardell | Full Text:PDF | | GTID:1462390011989183 | Subject:Geotechnology | | Abstract/Summary: | PDF Full Text Request | | Movement of ground that has liquefied because of earthquake shaking can cause considerable damage to buildings, bridges, pipelines, and port facilities. Lateral spreading is one form of earthquake-induced ground deformation that has caused extensive damage in past earthquakes. The methods available to predict the magnitude of lateral displacement include both analytical and empirical approaches. Bartlett and Youd (1992b) and Youd et al. (1999) applied multiple linear regression techniques to a database of documented case histories of lateral spread displacements to develop an empirically based method that is commonly used by practicing engineers. A limitation of Bartlett and Youd's procedure is that the independent variable used to characterize the thickness of the liquefiable layer, T15, does not consider the mechanics of liquefaction.; We developed a multiple linear regression equation for estimating lateral spread displacements from the case history database where the variable characterizing the soil profile was based on the mechanics of the liquefaction process. We investigated the influence of various soil profile properties on lateral spread displacements with the use of a nonlinear, effective stress-based ground response analysis program that included a state-of-the-art constitutive model to capture the contractive-dilative behavior of liquefied soil. The influential properties were combined into a single soil profile variable that can be readily computed from a typical exploration log. After formulating the soil profile variable, the ground response program was used to generate a database of more than 5,000 lateral spread displacement values. Regression analyses on this large database led to a preliminary form for the predictive equation that was subsequently refined with regression analyses on the available case history database. The result is a predictive equation developed by regression analyses on the database of lateral spread case histories where the variable used to characterize the soil profile accounts for the mechanics of the liquefaction process. | | Keywords/Search Tags: | Lateral spread, Soil profile, Variable, Ground, Case, Used | PDF Full Text Request | Related items |
| |
|