SEISMIC RESPONSE AND DAMAGE OF RETAINING STRUCTURES

EUGENE MARCIANO, Purdue University

Abstract

In the event of an earthquake, it is common that many retaining structures experience damage. A number of reconnaissance reports describe the condition of retaining walls and bridge abutments shortly after the occurrence of an earthquake. Methods have been developed to compute the movements of the wall. Their results indicate that it is generally impractical to design a retaining structure to sustain the peak seismic load without yielding. Designing for a permissible level of wall movement is economically more feasible. Observations and results of the reconnaissance reports and the experimental investigations support this finding. In a number of recent attempts for computing wall movements, the retaining structure is assumed to slide along a rigid base without tilting. The backfill is treated as either a rigid plastic Mohr-Coulomb granular material or as an elastic material having moduli varying from point to point in accordance with the level of strain occurring at each point. However, real retaining structures have been observed to tilt and are often founded on soil deposits that are compressible. In addition, the stiffness of the backfill material and the foundation material varies with time as well as location. In order to account for tilting, the nature of the foundation soil, and the variation of soil stiffness with time, a dynamic soil-structure-interaction model for free standing rigid retaining structures has been developed. The equations of motion are solved by explicit numerical integration. The stiffness of the soil at every point in the system is adjusted to be compatible with the level of strain at the end of each time step. A nonlinear, constitutive law is used to relate the soil stiffness to the stress-strain state. To verify the numerical model, an experimental investigation was undertaken involving shaking table tests. A rigid cantilever wall was used. Observed tilting of the retaining structure and deformation of the foundation material were significant in these tests. There was evidence of strain dependency in the soil response. Amplification of the table accelerations was also evident in the noted accelerations of the wall, which consequently affected the residual movements of the wall. (Abstract shortened with permission of author.)

Degree

Ph.D.

Subject Area

Civil engineering

COinS