Strong motion data obtained over the last decade from sites with instrumented structures and free-field accelerographs has provided an unprecedented opportunity to evaluate empirically the effects of soil-structure interaction (SSI) on the seismic response of structures. Strong motion data were gathered for 58 sites encompassing a wide range of structural systems, geotechnical conditions, and ground shaking levels. System identification analyses were employed with these records to quantify the effects of inertial interaction on modal parameters of structures. Simple indices of free-field and foundation-level ground motions were also compared. From these results, the conditions under which significant SSI effects occur were identified, and simplified analytical techniques for predicting these effects were calibrated.
For each site, system identification analyses were used to evaluate first-mode periods and damping ratios for a flexible-base case which incorporates SSI effects, and a fixed-base case in which only the structural flexibility is represented. Inertial interaction effects were evaluated from variations between fixed- and flexible-base parameters (i.e. the lengthening of first-mode fixed-base period due to foundation translation and rocking, and the damping attributable to foundation-soil interaction). These inertial interaction effects were found to be significant at some sites (e.g. period lengthening ratios of 4, and 30% foundation damping), and negligible at others (no period lengthening and zero foundation damping).
Analytical formulations similar to procedures in contemporary building codes were used to predict inertial interaction effects at the sites for comparison with the “empirical” results. A collective examination of the empirical and predicted results revealed a pronounced influence of structure-to-soil stiffness ratio on inertial interaction, as well as secondary influences from structure aspect ratio and foundation embedment ratio, type, shape, and non-rigidity. The analytical predictions were generally found to be reasonably accurate, with some limitations for deeply embedded and long-period structures.
Full List of PEER Reports: click here.