Experimental investigation of the seismic performance of caisson foundations supporting bridge piers

Abstract

Allowing the transitory attainment of bearing capacity of caisson foundations supporting bridge piers during strong seismic events can lead to substantial optimisation in their design and major cost savings. If the approach of Capacity Design is applied to geotechnical systems, the temporary triggering of plastic mechanisms may be permitted if the resulting permanent displacements are smaller than given threshold values. To validate this design approach, the seismic performance of caisson foundations was assessed through dynamic centrifuge testing on reduced-scale models. This paper presents the results of two tests in which a caisson-pier-deck system was embedded in a typical alluvial deposit and subjected to a series of earthquakes of different intensities. The caissons were founded on soft and very soft clay, to either avoid or induce the attainment of plastic soil behaviour under the same seismic inputs. It is shown that both yielding and failure of the layer of very soft clay limit inertial forces transmitted to the superstructure, validating the design approach and some useful empirical relations available in the literature. On the other hand, inelastic soil behaviour implies accumulation of permanent rotation and settlement of the system, which must be carefully evaluated to check for fulfilment of performance requirements.

Numerical Back-Analysis of Dynamic Centrifuge Tests on the Seismic Behaviour of Caisson Foundations Supporting Bridge Piers

“…where meff = 270.4 Mg is the effective mass related to the first mode of the pier [3], while the relative settlement wr was calculated as…”

“…Figure 1: Schematic layout of the problem at prototype scale (modified from [8]). For simplicity, the deck is assumed to be reproduced by a tip mass of ms = 194.40 Mg.…”

“…In this paper, reference to the Adana record is only made for the sake of brevity, which was characterised by a peak horizontal acceleration amax, inp = 0.133g (moderate intensity). (modified from [8]).…”

“…The soil mechanical parameters were calibrated against those obtained in the centrifuge tests reported in Gaudio et al [3]. First, the unit weight of the clay, , was obtained to reproduce the in-flight vertical effective stress profile, 'v0, which was measured before applying the ground motions: here it is worth mentioning that the stress state was affected by some residual excess pore water pressures in the clay, u, as the duration of consolidation preceding the dynamic phase did not last enough (1 hour only) to make u dissipate completely.…”

“…In this study, the centrifuge experiment performed by Gaudio et al [3] is reproduced in the Finite Element (FE) code Plaxis 3D CE 2022 [7], where a caisson-pier-deck system embedded in a typical alluvial deposit with soft clay is subjected to a moderate-intensity earthquake. The comparison of the numerical results against the experimental ones shows that the main features ruling the seismic performance of the system may still be captured if a proper calibration of soil constitutive parameters is carried out.…”