Download (4MB) | Preview


The protection of the building from seismic events is a fundamental phase in the structures design should be introduced to avoid the loss of lives especially when it occurs in developing countries. This natural calamity produces social and economic consequences because a lot of people are killed by the collapse of brittle heavy unreinforced masonry or poorly constructed concrete buildings. The engineers can use in their professional practice seismic isolation or energy dissipation devices to prevent these disasters. The first ones are elements integral for the stability, the second ones are elements not forming part of the gravity frame system. It is spread to protect important or special structures but there is an increasing interest for using these devices in houses, schools and hospitals, especially in those countries with a large risk of earthquakes. The problem is to simplify the seismic design in order to propagate this project philosophy in the professional practice. An example of seismic isolation analysed within this research is the intervention on the “Santuario delle Madonna delle Lacrime” in Siracusa, Italy. The seismic retrofit was made substituting the bearings supporting the impressive dome with sliding seismic isolators equipped with elasto-plastic dissipators. Each old bearing allowed the geometrical variations of the diameter of the base ring supported the cover due to thermal and tensional variations inside itself, while the displacements in tangential direction were prevented. The new anti-seismic devices, installed between the 22 columns of the structure and the truncated-conical dome during the raising and lowering phases, are unidirectional bearings including elasto-plastic dissipators with “moon’s sickle” shape, able to transmit the horizontal seismic action on the dome to the columns through their elasto-plastic movement. The elastic behaviour of a “moon’s sickle” element up to the achievement of the steel yield stress in the most stressed point was analytically examined, in order to compute the elastic stiffness that approximately corresponds to the one experimentally observed. A Finite Element Structural Analysis Program has been used to construct the simplified and the complete numerical model of the structure, able to simulate its real behaviour. In the first one model, the dome of the Sanctuary has been assumed as a rigid body supported on 22 r.c. piers uniformly distributed along the circular perimeter of the Upper Temple’s plan. An analytical model was worked up and compared with the model developed through SAP-2000 software. The seismic input for the numerical analyses is represented by 7 couples of artificial accelerograms compatible with the elastic response spectrum defined by the new code, (Ministerial Decree of 14 January 2008, G.U. n. 29 del 4.02.2008 suppl. ord. n° 30) and for each accelerogram a duration of 26s has been assumed. It was therefore decided to reactivate the monitoring system, which contained some breakdown elements due to the default of maintenance, through an intervention of overtime maintenance and adaptation to the new constraint scheme of the dome, in order to finally start the operations of monitoring and continuous control of the construction. This structure has been recently included among those of the Italian network of buildings and bridges permanently monitored by the Italian Department of Civil Protection of the Seismic Observatory of Structures (OSS). The energy dissipation study has been carried out with an extensive set of dynamic experimental tests, named JetPacs - Joint Experimental Testing on Passive and semi-Active Control Systems - within the topics no.7 of the ReLuis Project (University Network of Seismic Engineering Laboratories). The analysis have been performed by using a 2:3 scaled steel braced frame, available at the Structural Engineering Laboratory of the University of Basilicata in Potenza, Italy. During the experimental campaign, the structural model was subjected to three different sets of natural or artificial earthquakes, compatible with the response spectra of the Eurocode 8 and of Italian seismic code (OPCM 3431, 2005) for soil type A, B and D. The dissipation systems, developed with different materials and technologies, consist of six different types of passive or semi-active energy dissipating devices with different behaviours. The JETPACS mock-up model is a two storeys one-bay steel frame with composite steel-reinforced concrete slabs. It is well known that the efficacy of semi-active devices in controlling the dynamic response of a structure increases with the increase of the ratio between the first vibration period of itself and the time reactivity of the device. In order to elongate the vibration periods of the test frame, a modified symmetrical configuration has been obtained by adding four concrete blocks on each floor. Instead the efficacy of passive and semi-active energy dissipating devices in controlling the torsional behavior was been considered with only two additional concrete blocks on both the first and the second floors, creating eccentricity with respect to the mass center. Therefore the model has been experimentally analyzed in three different configurations namely: i) bare frame without any additional mass, designated as CB; ii) frame with four additional concrete blocks at first and second floors close to each corner, designated as CS; iii) frame with two additional concrete blocks on first and second floors placed eccentric with respect to mass center, designated as CN. Based on the detailed description of the JETPACS Mock-up model, attempt has been made to closely simulate the test specimen using the SAP-2000 software to match the experimental results of the dynamic characterization tests conducted at Structural Engineering Laboratory of the University of Basilicata. An uniform increase in thickness of 17 mm in the reinforced concrete slab has been considered, which is equal a certain increase of mass for each floor. The dynamic characteristics of the analytical model strongly depend on the extent of connectivity between the floor beam and the reinforced concrete slab. It was decided to completely ignore the rigidity between the floor beams and r.c. slab in the further analysis because during model fabrication, connections were quite weak and made just to support the vertical loads. The most common devices used for isolated buildings are multilayered laminated rubber bearings. They can be constituted by dowelled shear connectors or held in place by recessed plate connections. At the Earthquake Engineering Research Center of the University of California in Berkeley, under the guide of Prof. James M. Kelly, a study on the buckling and roll-out instability behaviour of non-bolted bearings under lateral and vertical load was worked up. The hypothesis that the onset of instability under lateral displacement is the critical pressure pcrit applied to the reduced area Ar was adopted to analyse the former aspects. This methodology rises observing that a large number of smaller bearings is less expensive than a smaller number of large bearings with variable sizes that need to be designed for different column loads. In fact the idea is that it is possible to adjust to the variable column loads by using one, two, three, four or five bearings under each column. The only question of concern is that of the stability of a set of bearings as compared to a single bearing with the same horizontal stiffness. Two case studies, in which the replacement of five small bearing with a single big bearing, have been examined. The bearings have been subjected to the downward displacement of the top due to horizontal displacement and vertical load applied. This shortening is fundamental in the design process of the bearing itself and can be considered by the buckling analysis. Three different bearing configurations under vertical loads and lateral displacements have been studied and the results compared with a numerical model available at UCB to simulate the real behaviour. The shape of the isolators usually is circular, rectangular or long strip. The latter are often used in buildings with masonry walls. The buckling and the roll-out displacement for an infinite strip bearing with width 2b was analyzed and it has been defined the ratio f, between the load and the critical load applied to the reduced area, to calculate the roll-out. New formulas have been elaborated to design the non-bolted bearings in order to adopt them in a future seismic code updating.

Item Type: Tesi di dottorato
Uncontrolled Keywords: Seismic isolation, Energy dissipation, Finite Element model, buckling and roll-out on rubber bearings
Depositing User: Stefania Castanò
Date Deposited: 10 Nov 2009 08:59
Last Modified: 30 Apr 2014 19:35

Actions (login required)

View Item View Item