Giovanni Biondi

Experimental study in the shaking table of the input motion characteristics in the dynamic SSI of a SDOF model

In conventional seismic design the capacity of the system is generally exploited only at the superstructure level. However, soil non-linearity as well as soil-foundation interface non-linearity can be crucial in the seismic response of structures. The results of tests performed on physical models allow the main aspects of these interaction mechanisms to be identified and also provide a benchmark for subsequent theoretical or numerical analyses. The present paper deals with two shaking table tests performed at the University of Bristol’s EERC laboratory. The tests were performed on a physical model consisting of a Leighton Buzzard sand deposit and a one-storey steel model structure. Some of the test results are presented and discussed in terms of acceleration and displacement responses. Both time- and frequency-domain representations were adopted to highlight the influence of the frequency and amplitude of the input motion on the coupled and/or uncoupled response of the tested soil-structure system, as well as the effect of soil non linear behaviour.

Displacement versus pseudo-static evaluation of the seismic performance of sliding retaining walls

Pseudo-static seismic analysis of retaining walls requires the selection of an equivalent seismic coefficient synthetically representing the effects of the transient seismic actions on the soil-wall system. In this paper, a rational criterion for the selection of the equivalent seismic coefficient is proposed with reference to sliding retaining walls. In the proposed approach earthquake-induced permanent displacements are assumed as a suitable parameter to assess the seismic performance and an alternative definition of the wall safety factor is introduced comparing expected and limit values of permanent displacements. Using a simplified displacement prediction model it is shown that, for a given design earthquake, reliable values of the equivalent seismic coefficient should depend on all the factors affecting the stability condition of the soil-wall system and on a threshold value of permanent displacement related to a given ultimate or serviceability limit state. To achieve a match between the results of the pseudo-static and of the displacement-based analysis, the proposed procedure detects the value of the equivalent seismic coefficient for which the two approaches provide the same factor of safety. Thus, without necessarily carrying out a displacement analysis, a measure of the safety condition of a soil-wall system consistent with the actual seismic performance may be achieved through an equivalent pseudo-static analysis.

SOME EXPERIMETNAL EVIDENCES ON DYNAMIC SOIL-STRUCTURE INTERACTION

Abstract. Due to dynamic soil-structure interaction (DSSI), the foundation soil experiences an additional motion which is added to its free-field response and DSSI is ruled by the soil compliance and by the vibration of the structure. In this framework soil non-linear behavior, as well as soil-foundation interface non-linearity, can represent crucial aspects and may govern the seismic response of the overall soil-structure system. Experimental tests on physical models allow identifying the interaction mechanism and also provide a benchmark for theoretical and/or numerical analyses. The present paper describes some experimental evidences concerning dynamic soil-structure interaction. Specifically, the paper deals with the results of two shaking table tests performed on a physical model consisting of a one-storey steel frame and a sand deposit. The steel frame (1.3 x 0.95 x 1.3 m) represents a 1:6 scaled model of a one-storey reinforced concrete building prototype. To reproduce the effects of the foundation soil on DSSI, a special laminar box (5m x 1m x 1.2m) was placed over the table, fixed to it and then filled with dry Leighton Buzzard sand up to a depth of 90 cm. During the tests sinedwell excitations, with different acceleration amplitudes and frequencies, were applied as input motions and the overall model response was monitored through a large set of accelerometers and displacement transducers. Selected experimental results are presented and discussed in the paper using both timeand frequency-domain representations of the acceleration responses. The obtained results highlight the influence of the frequency and of amplitude of the input motion on the coupled and/or uncoupled response of the considered soil-structure system. In some cases an uplifting of the foundation was clearly observed during the tests and represented a natural isolation for the system; accordingly, the accelerations recorded in the soil underneath the foundation are not completely transmitted.