Maria Rota

Shaking Table Test of a Strengthened Full-Scale Stone Masonry Building with Flexible Diaphragms

An extensive experimental campaign has been conducted at EUCENTRE to understand the dynamic behavior of historic stone masonry structures and evaluate the seismic performance of selected strengthening strategies, aimed at improving wall-to-floor connections and in-plane diaphragm stiffness. Shaking table tests were performed of full-scale two-storey buildings in undressed double-leaf stone masonry with timber floor and roof. A first prototype (Building 1), representing a vulnerable building without antiseismic detailing and devices, was tested showing a response characterized by in-plane distortion of the flexible diaphragms and local out-of-plane failure mechanisms. In Building 2 the wall-to-diaphragm connections were improved, providing only a moderate in-plane stiffening of the wooden diaphragms. When subjected to shake-table testing, the strengthened building showed a global type of structural response without the occurrence of out-of-plane mechanisms. In the present paper the strengthening interventions on Building 2 are described, and the results obtained during the dynamic tests are illustrated, focusing on the response of the structure, the evolution of damage mechanisms during the tests, in comparison to the seismic performance of the first unstrengthened reference prototype response. The improvement of the connections proved to be very effective, increasing significantly the seismic capacity of Building 2 with respect to Building 1.

ASCONA: Automated Selection of COmpatible Natural Accelerograms

This paper describes an automated procedure for selecting and scaling real spectrum-compatible records. The methodology allows one to choose from a predefined database, assembled from accredited strong-motion accelerometric data banks, real records satisfying properly defined seismological constraints with the additional requirement of spectrum-compatibility with a reference spectrum in a specified period range. Among the different sets of records satisfying these constraints, the user can specify the desired one, based on additional requirements (e.g., limited scaling factors). The proposed algorithm allows one to select records compatible with either an acceleration or a displacement response spectrum.

The Effect of Stiffened Floor and Roof Diaphragms on the Experimental Seismic Response of a Full-Scale Unreinforced Stone Masonry Building

An extensive experimental program was carried out at EUCENTRE, within a research project on the evaluation and reduction of the seismic vulnerability of stone masonry structures. The main part of the experimental program has been devoted to the shaking table tests on three full-scale, two-story, single-room prototype buildings made of undressed double-leaf stone masonry. The first building tested was representative of existing unreinforced stone masonry structures with flexible wooden diaphragms, without any specific anti-seismic design nor detailing. In the second and third buildings, strengthening interventions were simulated on structures theoretically identical to the first one, improving wall-to-floor and wall-to-roof connections and increasing diaphragm stiffness. In particular, in the third specimen, steel and r.c. ring beams were used to improve the diaphragm connection to the walls and collaborating r.c. slab and multi-layer plywood panels were used to stiffen floor and roof diaphragms, respectively. This article describes the strengthening interventions applied to the third building prototype and presents the experimental results obtained during the shaking table tests. The results obtained permitted the calibration of a macroelement model representative of the nonlinear behavior of the structure.

Evaluation of Uncertainties in the Seismic Assessment of Existing Masonry Buildings

The assessment of existing masonry buildings is an important issue in earthquake prone countries like Italy. The current Italian building code, which adopts the approach proposed by Eurocode 8, includes an assessment procedure based on the use of confidence factors, whose values depend on the level of knowledge of the structure. These factors are intended to take into account all possible uncertainties related to the incomplete knowledge of the structure. This article investigates the reliability of the code-based procedure for the assessment of existing masonry buildings and pinpoints some problematic aspects. The approach followed is the simulation of the entire code-based assessment procedure, with the flow of decisions that an engineer would face in the assessment of an existing building schematized in the form of a logic tree. The proposed simulated procedure accounts for different sources of epistemic uncertainty like the selection of the level of knowledge, uncertainty in the results and location of in-situ tests, identification of several structural details, etc. A Monte Carlo procedure allows the simulation of a large number of random assessments intended to be performed by different virtual engineers. The results are then compared with those coming from the assessment of the “perfectly known” structure, used as a benchmark, providing an estimate of the validity of the codified assessment methodology.

Mesozonation of the Italian territory for the definition of real spectrum-compatible accelerograms

The Italian building code defines the seismic action in terms of elastic acceleration response spectra derived from the results of a probabilistic seismic hazard study performed for the whole national territory. This representation of the seismic input is insufficient for several situations (e.g. analysis of geotechnical systems or time-history analyses of structures), for which the seismic input needs to be specified in terms of accelerograms. This work illustrates a methodology for the seismic mesozonation of the Italian territory, with the aim of defining suites of 7 real accelerograms recorded at outcropping rock sites with flat topographic conditions and, most importantly, compatible with the elastic acceleration response spectrum defined by the Italian building code at any location in Italy. These accelerograms do not require any correction and can be directly used for nonlinear dynamic analyses of structures and geotechnical systems. The mesozonation is based on identification of groups of spectra with similar characteristics and shape. For each of these groups, a parent spectrum is defined and used for selecting real spectrum-compatible records. Limited linear scaling is then applied to these accelerograms to make them compatible with all the response spectra of the group. The results of this work for the 475-years return period are accessible through the SEISM-HOME Web-GIS (www.eucentre.it/seismhome.html) providing, for any site in Italy, a suite of 7 real accelerograms spectrum-compatible, on average, with the acceleration response spectrum prescribed by the Italian building code. SEISM-HOME is a useful tool for practitioners needing ready-to-use time-histories for seismic analyses.

Identification of Suitable Limit States from Nonlinear Dynamic Analyses of Masonry Structures

Performance-based earthquake engineering, developed over the last decades for the design and assessment of other structures, can also be applied for masonry structures if the particularities of masonry are incorporated into the procedure. According to this methodology, structural performance can be assessed according to damage states which are identified through displacement/damage indicators. While various methods for the identification of limit states from the results of nonlinear static analyses exist, the identification of damage states from the results of nonlinear dynamic analyses is still uncertain. This article investigates a number of criteria allowing to identify the attainment of significant limit states from the results of time history analyses, in terms of appropriately identified response quantities. These criteria are applied to five building prototypes and their results are compared. A comparison with the limit states derived from nonlinear static analyses is also made.

Typological Seismic Risk Maps for Italy

This paper describes the methodology followed to derive typological seismic risk maps for Italy and then presents the results. In its classical definition, seismic risk is obtained from the convolution of hazard, vulnerability and exposure. Due to the absence of reliable data on exposure for the entire Italian territory, this study proposes typological seismic risk maps, obtained by simply convolving hazard and vulnerability for several building typologies characteristic of the Italian building stock. A specific hazard study in terms of PGA has been carried out. The results have then been convolved with empirical typological fragility curves, that were derived from data collected during post-earthquake surveys after the main Italian events of the last 30 years. Useful applications can be found for the typological seismic risk maps, both for risk mitigation strategies and for purely economical evaluations (e.g., insurance and reinsurance studies).

Consideration of modelling uncertainties in the seismic assessment of masonry buildings by equivalent-frame approach

Abstract The choice of modelling strategy and analysis options has a significant influence on the results of the seismic assessment of existing buildings and therefore it is very important to have an idea of the dispersion in the results due to different hypotheses regarding the structural model. This paper concentrates on pushover analysis, considered as the reference method currently adopted by engineers for the seismic assessment of existing masonry buildings, and on the equivalent-frame macro-element approach, assumed to be a satisfactory compromise between computational effort and accuracy in the results. A logic tree approach is used to treat the different considered options, including the definition of the geometry of the equivalent frame, the distribution of loads among the masonry piers and on the horizontal diaphragms, the degree of coupling between orthogonal walls, the definition of the cracked stiffness of structural elements and the modelling of masonry spandrels. By assigning a value of probability to each end branch of the tree, the distribution of the peak ground acceleration corresponding to the selected limit states can be obtained and, from this distribution, a quantitative estimate (in probabilistic terms) of the effect of modelling uncertainties on the seismic response of masonry structures is derived.

ISSUES ON THE USE OF TIME-HISTORY ANALYSIS FOR THE DESIGN AND ASSESSMENT OF MASONRY STRUCTURES

Nonlinear dynamic analysis is widely recognized as the most accurate analysis technique for the design and assessment of structures. However, its use is still not common in the everyday engineering practice, mainly due to various issues concerning its performance and the interpretation of its results. A first issue includes the limited availability of computer programs that allow the performance of time history analysis, especially for the case of masonry structures, although some software are currently available to this aim. Another difficulty is related with the selection of appropriate input ground motion records to be used for the analysis. Real records are well known to be a preferable choice with respect to artificial or synthetic ground motions, but the limited availability of real records often requires scaling them, with all the concerns associated with this operation. Also, a proper selection of seismic input requires some level of expertise, which is not so common in the professional field. A third problem regards the difficulty in the interpretation of the results of nonlinear dynamic analysis in terms of performance limits. The definition of significant limit states in relation to the results of nonlinear dynamic analyses is still indeed a very open problem and it is related to the fact that the achievement of a local limit condition (e.g. failure of a pier element) would not adequately represent the overall damage state of the building. Therefore there is the need to find a definition of limit states describing the global building performance, i.e. taking into account not only the peak concentrated damage in a single element, but also the diffusion of damage through the different structural elements and the evolution of the global collapse mechanism. This would also allow for a rational implementation of the qualitative definition of damage states commonly adopted in performance

Seismic assessment of masonry buildings accounting for limited knowledge on materials by Bayesian updating

based earthquake engineering.