Di Domenico, Mariano (2018) Out-of-plane seismic response and modelling of unreinforced masonry infill walls. [Tesi di dottorato]

Anteprima

Testo didomenico_mariano_31.pdf Download (23MB) | Anteprima

Abstract

Past and recent earthquakes showed that the seismic performance of buildings is strongly influenced by the presence and contribution of unreinforced masonry (URM) infills, which are usually considered as non-structural elements. Such enclosures are used – especially in Mediterranean countries – to provide buildings with thermic, visual and acoustic insulation. On one hand, URM infills can stand significant lateral loads and, so, they contribute to the lateral strength capacity of structures. In addition, they are provided with a high in-plane stiffness. For this reason, the assessment of a construction modelled as bare frame can yield to a significant underestimation of its lateral strength and stiffness. On the other hand, it is well-known that the high force demand that URM enclosures attract and then transfer to the confining elements can yield to unexpected failures of structural members designed without accounting for infills’ presence. For example, Reinforced Concrete (RC) columns (and beams, potentially) not designed addressing seismic and capacity design provisions sometimes exhibit brittle failures during strong earthquakes due to the so-called “frame-infill interaction”, i.e., due to the shear forces transferred by infills and not considered in the design. Moreover, the absence of infills at a certain storey of a building (typically, the first) produces a non-negligible stiffness variation of the structure lateral stiffness along its height, leading, in this way, to potential peaks of inelastic demand at that storey yielding to a sidesway collapse due to a soft-storey mechanism. In addition, infills’ damaging due to IP actions and their repair or refurbishment produces most of the financial losses consequent to earthquakes. In other words, neglecting infills’ presence and their contribution to the seismic response of structures can be both conservative and unconservative. For these reasons, the interest in the characterization of the seismic response of URM infills has significantly grown in the engineering and research community in the last decades. It should be noted that these bi-dimensional non-structural elements are subjected to the seismic action both in the in-plane (IP) and in the out-of-plane (OOP) direction. The expulsion or overturning from the confining frame due to OOP actions of URM infills is potentially highly detrimental for human life safety and amplifies the economic losses consequent to earthquakes. The OOP collapse of URM infills is promoted by the damage due to IP actions, which can reduce their OOP strength, stiffness and displacement capacity. This phenomenon is called IP/OOP interaction. This PhD thesis is dedicated to the characterization and modelling of the OOP behaviour of URM infills and to the study of the effects of the IP/OOP interaction both at the level of the single (non-structural) component and at the level of the infilled structure seismic performance. Chapter I is dedicated to the existing literature concerning this issue and investigating the definition of the OOP strength, stiffness and displacement capacity of URM infills. In addition, existing formulation for the prediction and reproduction of the IP/OOP interaction effects are addressed. Finally, existing URM infills’ modelling strategies accounting for their OOP behaviour and for the IP/OOP interaction effects are described in detail. Chapter II constitute the second part of the previous literature recall, as it is dedicated to a detailed description of the experimental tests carried out in the past to investigate the OOP behaviour of URM infills and the IP/OOP interaction effects. It is observed that the experimental database allowing evaluating the effectiveness and robustness of literature formulations and models described in Chapter I is extremely poor. For this reason, a comprehensive and extended experimental program has been carried out at the Department of Structures for Engineering and Architecture of University of Naples Federico II. The experimental program main aim is the characterization of the effects of the panel height-to-thickness slenderness ratio, of the boundary conditions at edges and of the IP/OOP interaction on the OOP strength, stiffness and displacement capacity of URM infills. A total of fifteen tests has been carried out to enrich the available experimental database. Chapter III is dedicated to a detailed description of the experimental program and of its results. In Chapter IV, the experimental database collected in Chapter II and III is analysed and discussed, in order to compare the prediction of literature formulations and models aimed at assessing the OOP response of URM infills and/or its significant parameters, such as the force at first macro-cracking and at maximum, as well as their secant stiffness at the first macro-cracking and at maximum and the displacement capacity/ductility. The predicting capacity of the available IP/OOP interaction models is assessed, too. This comparison is aimed at evaluating the effectiveness of the available models for the prediction of the OOP response of both IP-undamaged and IP-damaged URM infills. Based on the results of this comparison, original and mechanical based proposals are described for a robust and effective modelling of URM infills’ OOP response. In addition, empirical formulation for the prediction of the IP/OOP interaction are proposed. With Chapter IV, the characterization of the OOP behaviour of the single panel, which is the first part of this thesis, is completed. Chapter V is dedicated to a simple state-of-the-art concerning the current provisions given by international technical codes and standards for the assessment of URM infills safety with respect to OOP seismic demands. More specifically, demand and capacity models provided by codes are described and discussed. This is preliminary to the assessment of the seismic performance of RC buildings accounting for the OOP response of infills and for the IP/OOP interaction effects, which is the second part of this thesis. To this aim, a set of sixteen case-study buildings has been designed according to Eurocodes’ provisions. The case-study buildings are described and commented in detail in Chapter VI. The case-study buildings are different for the number of storeys, which is equal to 2, 4, 6 or 8, and for the design peak ground acceleration (PGA) at Life Safety Limit State. In Chapter VII, the case-study buildings described in the previous section are used to assess the PGA at the first OOP collapse of different infill layouts in a non-linear static framework by using both a simplified “Designer (code-based) Approach” and a refined “Reference Approach”. Only the least accounts, in evaluating the OOP force demand and capacity of infills, for the structural nonlinearity as well as for the IP/OOP interaction effects. The dependence of such the PGA capacity with respect to the first OOP collapse on the number of storeys and on the design PGA is discussed. In addition, the PGA at the first OOP collapse is compared with the design PGA of the case-study buildings as well as to the PGA corresponding to their conventional structural collapse. It is shown that weak infills in mid- and high-rise buildings can collapse for OOP actions and due to the IP/OOP interaction effects at PGA demand lower than the PGA at structural collapse or even than the design PGA. In addition, simplified criteria to evaluate, based on infills geometric and mechanical properties, if the OOP safety check is necessary or not are presented. In Chapter VIII, the seismic performance of the case-study buildings accounting and not accounting for the IP/OOP interaction effects on URM infills is assessed by means of non-linear incremental dynamic analysis. Also in this case, the overall capacity of buildings with respect to the first OOP collapse is investigated. In addition, the OOP behaviour factor and effective stiffness of URM infills accounting for the IP/OOP interaction effects is evaluated. Such values can be used for a simplified OOP safety check of URM infills in a linear elastic framework. In the appendix section, some theoretical considerations and experimental data supporting the discussions above proposed are reported in detail.

Downloads

Actions (login required)

Modifica documento