Search

The interaction between soil, foundation and structure is a complex phenomenon in which several mechanisms are involved. Usually the soil structure interaction (SSI) is considered formed by two main mechanisms: the kinematic and the inertial interaction. Theoretical studies on the subject abound in the literature show that the effects of SSI on the earthquake response of structures included primarily the period elongation and the increase of equivalent viscous damping. Nonetheless, the experimental validations of such studies are still scarce. Such validations can be efficiently investigated with physical scaled model in 1-g or n-g devices. In the present paper the SSI is investigated by utilizing the set of data selected from a comprehensive experimental test program carried out on 1-g shaking table tests. The scaled model adopted in the shaking table tests comprises a group of five piles embedded in a by-layer deposit with or without a single degree of freedom connected to the piles with and without a rigid connection between them. The present work focused on the response of the scaled model when subjected to the strong motion recorded at Tolmezzo Station during the 1976 Friuli (Italy) earthquake. The experimental outcomes are compared with analytical results derived from well-established theoretical formulations existing in the literature for both the period elongation and the variation of the damping ratios. It is found that: (i) the SSI effect are, as expected, more evident in the system without connection between piles and (ii) the theoretical formula are not able to predict in an accurate manner these quantities due to the hypotheses at their base.

The seismic soil-pile-structure interaction (SPSI) is a complex mechanism that is usually considered formed by the combination of kinematic and inertial interaction. Both mechanisms generate additional forces on the system related to the stiffness contrast between the soil and the foundation for the kinematic interaction and to the soil response around the foundation due to the inertial contribution of the superstructure for the so-called inertial interaction. While the mechanism of each of these contributions is clear, their combination is still under investigation due to number of parameters involved (i.e. dynamic characteristics of both system and input). An effective way to study this combination is the analysis of actual data on real structures. Due to the fact that these data are hard to find, usually the response of physical scaled models on 1-g and n-g devices are investigated. In this connection, this paper presents some results from an extensive 1-g shaking table testing activity. The scaled physical model is formed by a group of five piles embedded in a by-layer deposit of dry sands with an oscillator connected to the piles through different kind of foundation systems. More specifically, the attention is focused on both pile and structural response when the oscillator is connected to a small group of three piles by means of a stiff foundation. The analysis of the experimental data enhances the role of the resonance between the soil-structure system and the input waves in the general behavior of both structure and piles

An effective way to study the complex seismic soil-structure interaction phenomena is to investigate the response of physical scaled models in 1-g or n-g laboratory devices. The outcomes of an extensive experimental campaign carried out on scaled models by means of the shaking table of the Bristol Laboratory for Advanced Dynamics Engineering, University of Bristol, UK, are discussed in the present paper. The experimental model comprises an oscillator connected to a single or a group of piles embedded in a bi-layer deposit. Different pile head conditions, that is free head and fixed head, several dynamic properties of the structure, including different masses at the top of the single degree of freedom system, excited by various input motions, e.g. white noise, sinedwells and natural earthquake strong motions recorded in Italy, have been tested. In the present work, the modal dynamic response of the soil–pile–structure system is assessed in terms of period elongation and system damping ratio. Furthermore, the effects of oscillator mass and pile head conditions on soil–pile response have been highlighted, when the harmonic input motions are considered.

During earthquake ground shaking earth pressures on retaining structures can cyclically increase and decrease as a result of inertial forces applied to the walls and kinematic interactions between the stiff wall elements and surrounding soil. Limit equilibrium analysis imposes a pseudo-static inertial force to a soil wedge behind the wall (the mechanism behind the widely-used Mononobe-Okabe method), which is a poor analogy for either inertial or kinematic wall-soil interaction. Many basement walls and retaining structures are dominated by kinematic soil-structure interaction (SSI) effects arising from differences in displacement between the wall and the free-field soil. Kinematic SSI solutions are often formulated for uniform soil conditions, but the shear modulus of most soils is known to increase with mean effective stress, and therefore with depth. We examine the influence of vertical heterogeneity of shear modulus on kinematic SSI for rigid walls. An existing free-field displacement solution is presented first, followed by analysis of earth pressure increments using a Winkler assumption. Vertical heterogeneity is shown to reduce seismic earth pressures compared with a uniform soil case (for a given frequency and peak ground surface displacement) because free-field displacements are largest near the surface, where the soil is softest and Winkler stiffness is lowest. The proposed Winkler solution is then compared with an exact analytical solution for vertically heterogeneous soil over a rigid base and retained between two opposing rigid walls. The agreement is imperfect, but reasonable, with differences likely due to assumptions regarding the dynamic Winkler stiffness intensity.

The geotechnical systems are often studied analysing the response of scaled models in 1-g or n-g devices. A crucial aspect in these kind of models is the capability to read the real response of the system using different instrumentations. In this paper are described the main devices used for the interpretation of the dynamic response of a scaled geotechnical model tested at the Bristol Laboratory for Advanced Dynamics Engineering (BLADE) of the University of Bristol, UK, within the framework of the Seismic Engineering Research Infrastructures for European Synergies (SERIES), funded by the 7th FP of the European Commission. The model is formed by a group of five piles embedded in a bi-layer deposit with different pile head conditions, different dynamic properties of the superstructure (connected at the top of the pile group) and input motions. After some details about the whole experimental campaign and the devices used, the paper shows the main procedures adopted to evaluate the response of the whole system in terms of acceleration, displacement and stresses (only for piles).

The geotechnical systems are often studied analysing the response of scaled models in 1-g or n-g devices. A crucial aspect in these kind of models is the capability to read the real response of the system using different instrumentations. In this paper are described the main devices used for the interpretation of the dynamic response of a scaled geotechnical model tested at the Bristol Laboratory for Advanced Dynamics Engineering (BLADE) of the University of Bristol, UK, within the framework of the Seismic Engineering Research Infrastructures for European Synergies (SERIES), funded by the 7th FP of the European Commission. The model is formed by a group of five piles embedded in a bi-layer deposit with different pile head conditions, different dynamic properties of the superstructure (connected at the top of the pile group) and input motions. After some details about the whole experimental campaign and the devices used, the paper shows the main procedures adopted to evaluate the response of the whole system in terms of acceleration, displacement and stresses (only for piles).

Event Overview and Organization of Reconnaissance Activities,Overview of Reconnaissance Activities, Regional Faults and Seismic Source,Ground Motions, Landslides and Rockfalls, Performance of Building Structures and Damage Patterns,Performance of Bridge Structures, Other Infrastructure Systems, Lessons Learned and Next Steps

The present analytical work discusses the outcomes of a series of 1-g shaking table experimental tests that were carried out to validate numerical models formulated for kinematic and inertial interaction effects on pile-supported systems. Towards this aim, pile models in layered sand deposits were built in the laboratory; such models were subjected to several cyclic tests and an ensemble of earthquake loading. The piles were densely instrumented with accelerometers and strain gauges; therefore, earthquake response, including bending strains along their length, could be measured directly. Different configurations were considered for the shake-table tests; the latter configurations include free-head piles and single-degree-offreedom (SDOF) systems with short and long caps at foundation level. The experimental data have been assessed accurately to estimate the period elongation of the SDOF structures, if any. Additionally, comparisons between the soil free-field response derived experimentally and advanced numerical simulations are also included. The results of the analyses show that the period elongations of the SDOF structure caused by pile-soil-interactions may be significant, thus affecting the evaluation of structural response under earthquake loading. Implications on the assessment of existing structures and the design of new ones are discussed

Kinematic effects at the head of a flexible vertical pile embedded in a two-layer soil deposit are investigated by means of rigorous three-dimensional elastodynamic finite-element analyses. Both pile and soil are idealized as linearly viscoelastic materials, modelled by solid elements, without the restrictions associated with the use of strength-of-materials approximations. The system is analyzed by a time-Fourier approach in conjunction with a modal expansion in space. Constant viscous damping is considered for each natural mode, and an FFT algorithm is employed to switch from frequency to time domain and vice versa in natural or generalized coordinates. The scope of the paper is to: (a) elucidate the role of a number of key phenomena controlling the amplitude of kinematic bending moments at the pile head; (b) propose a simplified semianalytical formula for evaluating such moments; and (c) provide some remarks about the role of kinematic bending in the seismic design of pile foundations. The results of the study provide a new interpretation of the interplay between interface kinematic moments and corresponding head moments, as a function of layer thickness, pile-to-soil stiffness ratio, and stiffness contrast between the soil layers. In addition, the role of diameter in designing against kinematic action, with or without the presence of an inertial counterpart, is discussed.

The present analytical work discusses the outcomes of a series of 1-g shaking table experimental tests that were carried out to validate numerical models formulated for kinematic and inertial interaction effects on pile-supported systems. Towards this aim, pile models in layered sand deposits were built in the laboratory; such models were subjected to several cyclic tests and an ensemble of earthquake loading. The piles were densely instrumented with accelerometers and strain gauges; therefore, earthquake response, including bending strains along their length, could be measured directly. Different configurations were considered for the shake-table tests; the latter configurations include free-head piles and single-degree-of-freedom (SDOF) systems with short and long caps at foundation level. The experimental data have been assessed accurately to estimate the period elongation of the SDOF structures, if any. Additionally, comparisons between the soil free-field response derived experimentally and advanced numerical simulations are also included. The results of the analyses show that the period elongations of the SDOF structure caused by pile-soil-interactions may be significant, thus affecting the evaluation of structural response under earthquake loading. Implications on the assessment of existing structures and the design of new ones are discussed

The cyclic and seismic response of soil-pile-structure systems is assessed through comprehensive experimental tests performed on the 3mx3m shaking table of the Bristol Laboratory for Advanced Dynamics Engineering (BLADE), University of Bristol (UK). Such tests were carried out in the framework of the Seismic Engineering Research Infrastructures for European Synergies (SERIES). The present work discusses preliminary results of the cyclic response of a pile group in a bi-layered soil profile. The outcomes of the test results discussed hereafter emphasize that kinematic effects are of paramount importance for the seismic analysis and design of structural systems with piled foundations. Appropriate combination rules to account for both inertial and kinematic effects are deemed necessary. The experimental results derived with the shaking table tests will be employed to calibrate numerical models, which, in turn, will be utilized to perform comprehensive parametric analysis aimed at providing sound design rules to be implemented in next generation performance-based seismic codes of practice.

According to modern seismic codes such as Eurocode 8, pile foundations in earthquake-prone areas must resist two different, yet simultaneous bending actions resulting from kinematic and inertial interaction. Due to the different nature of the two demands, pile must resist seismic actions following different patters, thus leading to different design requirements. In this work, analytical solutions are presented to define maximum and a minimum pile diameters required to resist kinematic and inertial effects in an essentially elastic manner, respectively. It is shown that the range of admissible diameters decreases with decreasing soil stiffness and with increasing design acceleration, collapsing into a single admissible diameter for certain problem configurations. Regions where no pile diameter can guarantee elastic response during strong seismic shaking are identified.

During seismic shaking soil movements force piles to deform, resulting in a complex interplay between the two systems commonly referred to as “kinematic interaction”. This interaction generates internal forces along piles even in the absence of a superstructure. Current design practice usually takes into account only forces transmitted to the pile from the superstructure, thus neglecting kinematic interaction. Despite intense research in the topic, few contributions focus on kinematic effects at the pile head in the presence of a stiff pile cap restraining its rotation. The subject may be of importance,for the pile head is subjected both to relevant kinematic and inertial moments, the latter attenuating rapidly with depth. The paper presents the results of an extensive parametric study leading to a simplified formula for evaluating kinematic moments at the pile head in a two-layer soil profile.

The L’Aquila earthquake occurred on April 6 2009 at 03:32:39 local time. The earthquake (Mw=6.3) was located in the central Italy region of Abruzzo. Much of the damage occurred in the capital city of L’Aquila, a city of approximate population 73000, although many small villages in the surrounding region of the middle Aterno river valley were also significantly damaged. In the weeks following the earthquake, the Geo-Engineering Extreme Events Reconnaissance (GEER) international team, comprised of members from different European countries and the U.S., was assembled to provide post-earthquake field reconnaissance. The GEER team focused on the geological, seismological, and geotechnical engineering aspects of the event. We describe the principal seismological findings related to this earthquake including moment tensors of the main shock and two triggered events, the aftershock pattern and its variation with time, tectonic deformations associated with the main shock, surface fault rupture, and the inferred fault rupture plane. We describe damage patterns on a village-to-village scale and on a more local scale within the city of L’Aquila. In many cases the damage patterns imply site effects, as neighbouring villages on rock and soil had significantly different damage intensities (damage more pronounced on softer sediments). The April 6 mainshock was the best-recorded event to date in Italy. We present metadata related to the recording sites and then present preliminary comparisons of the data to GMPEs. Those comparisons support the notion of faster distance attenuation in Italy relative to the average for active regions as reflected in NGA GMPEs. Several incidents of ground failure are then discussed, including a number of rockfalls and minor landslides. Perhaps the most significant incidents of ground failure occurred at Lake Sinizzo, for which we describe a number of slumps and spreads around the lake perimeter. This is documented using field observations as well as LIDAR and bathymetric data.