Maria Polese

A simulated design procedure for the assessment of seismic capacity of existing reinforced concrete buildings

This paper presents the general criteria and implementation of an automatic procedure to evaluate the seismic capacity of existing reinforced concrete (RC) regular buildings. The method represents a useful tool in the framework of mechanical based vulnerability assessment methods. In particular, the seismic capacity is retrieved via pushover analyses on a lumped plasticity model for the building. Unlike recent approaches that rely on a single representative structural model for an entire building population, the proposed method allows virtually all the buildings of the population to be analysed in an automatic loop. With the aim of expediting and automatizing the analysis process, a dedicated software was implemented, whose main functions are: identifying possible structural systems compatible with regular building shapes of assigned dimensions and designing its elements in terms of cross-section and reinforcement; constructing the related nonlinear model and performing pushover analyses in order to determine synthetic capacity parameters useful for preliminary vulnerability assessment at a large scale. The software application and potentialities are shown in an example for the generic building of a class.

A MULTILEVEL APPROACH TO THE CAPACITY ASSESSMENT OF EXISTING RC BUILDINGS

In the present paper a rational mechanical based approach for the seismic capacity assessment of classes of buildings is presented and capacity curves in terms of ultimate strength and deformation capacities are derived. The proposed procedure allows the main parameters (morphologic and geometric configuration, mechanical properties etc.) to be chosen and their relative influence on the capacity of RC buildings to be evaluated. This evaluation is helpful, since it permits, in the framework of vulnerability assessment at a regional or sub-regional scale, to have a measure of the error that is related to the amount of the information available. Consequently, depending on the relative influence of parameters on response, and also considering the different type of parameters and related availability (by public databases for low order parameters, or by field survey for high order ones), a multilevel building class specification in function of input variables is proposed.

Vulnerability Analysis for Gravity Load Designed RC Buildings in Naples – Italy

Damage scenario and risk analyses are helpful tools for the local administrators for mitigation of potential earthquake losses at the urban-regional level. One of the main issues in developing such scenarios is the choice of proper capacity functions expressing the effective seismic supply for the existing building classes and the convolution with demand in the so-called fragility analysis. This article, as a further implementation of a recently developed mechanical based procedure for class scale quantitative risk evaluation, presents the derivation of class representative capacity curves and the relative fragility curves for slight, moderate, extensive, and complete damage states as defined by the well-known HAZUS methodology. Starting from an extensive building survey of Arenella district in Naples, southern Italy, statistics on main model input parameters are obtained for selected building classes of existing and/or pre-code RC buildings. Accordingly, a number of building models is simulated designed and analyzed in order to determine building class capacity. Fragility curves are computed simulating the fraction of “failures” within a capacity spectrum method framework. These capacity and fragility curves have been used in a companion paper by Lang et al. [2008] for the computation of damage scenarios in Arenella.

Mechanism based assessment of damage-dependent fragility curves for RC building classes

Seismic fragility curves play a critical role in risk assessment because they represent the probability of attaining different damage states given the ground motion intensity. However, in case of repeated earthquakes, the seismic vulnerability of buildings varies due to damage and this shall be properly considered for the realistic estimate of evolving seismic risk during a seismic sequence. This paper presents a methodology for the assessment of damage-dependent fragility curves for existing reinforced concrete (RC) building classes. The derivation of fragility curves is based on a hybrid method that combines observational based “empirical” curves and mechanical based assessment of buildings’ residual capacity (REC), that is a measure of seismic capacity, which may be reduced due to seismic damage. A mechanism based procedure to assess building REC, and its variation for increasing ductility demand, is presented. Then, in order to be able to apply such procedure to RC building classes, a simulated design procedure is implemented that allows the automatic design of elements dimensions and reinforcement, as well as the characterization of the building nonlinear model needed for mechanism based analysis. Applying a Monte Carlo simulation procedure the geometric, structural and mechanical parameters, from which the design depends on, are varied according to representative distributions and populations of building models are generated and analyzed with mechanism based approach. The procedure is implemented for buildings that are prone to first storey mechanism and is applied for existing RC building classes in L’Aquila.

Reconstruction policies: explicitating the link of decisions thresholds to safety level and costs for RC buildings

In (FEMA 308 in Repair of earthquake damaged concrete and masonry wall buildings, Prepared by ATC for the Federal Emergency Management Agency, Washington DC, 1998) a general framework facilitating decisions on appropriate course of action (accept damage, restore, or upgrade) for damaged buildings after an earthquake was presented. Such Performance-Based Policy Framework relies on performance index of the building in its intact and damaged state and on the relative performance loss as significant indicators for repair and/or upgrade decisions; however, no specific guidance for the establishment of PL and IP thresholds governing damage acceptability was given. This paper proposes an improvement of the PBPF introduced in FEMA 308 by providing it with a set of performance thresholds that can be established based on a clear quantitative approach and presents a set of tools for its practical implementation with reference to Reinforced Concrete frame buildings. In addition to the Repair and Repair and Retrofit options, also the Demolish one is introduced, because it corresponds to a not infrequent decision in case of severely damaged buildings in European-Mediterranean regions. Criteria to establish decision thresholds IP and PL are proposed that allow to clearly connect them to reconstruction costs and probability of failure of the building in the intact and damaged state. Finally, a case study demonstrates the application of the hypothesized policy framework for a Municipality in southern Italy and investigates on the effects of varying decision thresholds on the total reconstruction costs and mean safety level for the considered building stock.

Simulation of Earthquake Ground Motion and Effects on Engineering Structures during the Preeruptive Phase of an Active Volcano

Several of the worlds active volcanoes are located near densely pop- ulated areas, and therefore the seismic hazard associated with preeruptive earthquake activity and its relation to the potential damage of engineering structures should be considered as a part of risk evaluation and management. This is true for Mt. Vesuvius volcano (southern Italy), where several hundred thousand people are exposed to volcanic and related seismic risks. This study investigates the effect of preeruptive seismic activity through the massive simulation of earthquake waveforms in the magnitude, location, and focal-mechanism ranges expected for Mt. Vesuvius seis- micity. Synthetics are processed to evaluate the characteristic strong-motion param- eters, which are useful for estimating the seismic damage to a built-up environment that would arise from both the maximum expected single event and the cumulative effect of a large number of small events. Synthetic and observed strong-motion pa- rameters for a selected set of recorded earthquakes are compared to validate the modeling approach. The scaling of the simulated peak ground acceleration (PGA) with distance appears to be influenced by earthquake depth, owing to the presence of a sharp velocity discontinuity at shallow depths underneath the Vesuvius area. On the other hand, the hysteretic energy spectrum, related to the plastic behavior of the structures, depends strongly on the b-parameter of the Gutenberg-Richter law (G- R). By varying the G-R law parameters across a reasonable expected range, we observe that the cumulative hysteretic energy is comparable to the values observed at near-source distances during the 1997 Umbria-Marche, Italy, event (M 5.8), which produced serious damage to buildings and infrastructure, although a significant PGA value was not recorded.

MECHANISM BASED ASSESSMENT OF DAMAGED BUILDING'S RESIDUAL CAPACITY

Seismic behavior of damaged buildings may be expressed as a function of their RE- sidual Capacity (REC). The residual capacity REC Sa is defined as the minimum spectral acce- leration (at the period T eq of the equivalent SDOF) corresponding to building collapse. When referring to peak ground acceleration a g as damaging intensity parameter, REC ag is defined as the minimum anchoring peak ground acceleration such as to determine building collapse. For a given spectral shape, REC ag corresponds to REC Sa scaled by the spectral amplification factor for T eq . REC Sa and REC ag , generally indicated as REC, lower with increasing damage level in buildings; hence REC may be very useful in estimating the post-seismic building safe- ty. In a recent work (1) it has been shown how it is possible to derive REC (REC Sa and REC ag ) through Pushover Analyses (PA), where a suitable modification of plastic hinges for damaged elements is applied. The applicability of PA for damaged structures is verified in (2) by com- parison of the PA results with those on nonlinear time-history analyses. On the other hand, it is unrealistic that in the aftermath of an earthquake, when the assessment of building safety has to be performed in an emergency situation, there would be time for the execution of de- tailed nonlinear analyses. Acknowledging the need for easier and faster evaluation tools, in (3) a simplified MEChanism based method (MEC) for evaluating the building REC was pre- liminary tested. The present work extends the comparison of the results (in terms of REC), that could be obtained by PA and MEC analyses, considering a number of Reinforced Con- crete (RC) frames building typologies. Moreover, by adopting the MEC approach, the possi- ble variation of REC as a function of seismic demand is investigated. The simplified method can be used to explore the possible ranges of REC variation for RC building classes consider- ing the anticipated mechanism formation after an earthquake; in addition, it could be used in the post-seismic phase for estimating the REC of damaged buildings having undergone identi- fiable plastic mechanisms and for fast assessment of damage-dependent collapse fragility curves.

Simplified assessment of maximum interstory drift for RC buildings with irregular infills distribution along the height

This paper investigates on the effect of story lateral stiffness variation on the maximum elastic interstory drift ratio (IDRmax) for existing reinforced concrete (RC) buildings. Several classes of existing gravity load designed RC buildings are obtained via a simulated design approach. The presence of infills in the perimeter frames as well as different opening percentages along the height are considered. A simplified elastic analysis is performed, adopting an equivalent multistory cantilever system to represent the stiffness variation along the buildings height. IDRmax is significantly influenced by the ratio of the lateral stiffness at the second and upper stories over the lateral stiffness of the first story. Such a ratio has been found dependent on a number of geometric and configuration factors, including the variation of the opening percentage ratio. Two regression formulas are proposed to estimate, given the spectral displacement at the fundamental period T, the roof displacement and IDRmax as a function of the building height and the opening percentage at first and upper stories. Suitable modification of the formulas is also introduced to account for possible cracking of RC elements and infill panels even at very low levels of lateral drift. These expressions could be used for the simplified evaluation of the expected drift demands for buildings of existing RC typologies. Finally, the proposed formulations are applied to a number of permanently monitored buildings, comparing the calculated IDRmax with the one resulting from record processing, obtaining a fair good agreement.

INFLUENCE FACTORS FOR THE ASSESSMENT OF MAXIMUM LATERAL SEISMIC DEFORMATIONS IN ITALIAN MULTISTOREY RC BUILDINGS

In the aftermath of low-to-moderate earthquakes, simplified methods allowing fast estimate of maximum lateral deformations experienced in multistory building are particularly useful when estimating nonstructural damage on a large building inventory. In particular, Miranda [1] introduced a simplified method that adopts an equivalent continuum elastic structure to estimate both maximum roof displacement and maximum interstorey drift ratio for a given response spectrum. Based on a closed-form solution, this method was further extended [2] to account for higher vibration modes, different distributions of lateral forces and non-uniform stiffness along the height. However, the method assumes that building properties will continuously vary along the height, assumption that is not generally valid for the Italian building stock due to the presence of knee beams and infill panels. This paper investigates on the effect of variation of infill properties on the elastic lateral response of infilled RC buildings, assessing the influence of different characteristics, such as geometric and material properties or the opening percentage, on IDRmax. Plane Multi-degree-of-freedom elastic models including explicit modeling of infill panels are adopted and the effect of local discontinuities is included, while linear time-history analyses are performed considering several input ground motions. The effect on the interstorey drift ratio of different opening percentages and infill arrangements, both at the same storey and along the height, are considered to account for discontinuities caused by the presence of infills. Further, the influence of different types and thicknesses of the infill panels, level connection to the surrounding frame is analyzed. 4161 Available online at www.eccomasproceedia.org Eccomas Proceedia COMPDYN (2017) 4161-4173

SIMPLIFIED ASSESSMENT OF EXPECTED SEISMIC LOSSES FOR AS BUILT AND RETROFITTED RC BUILDINGS

Seismic risk is a key factor influencing several important decision-making processes from performance-based design of new structures, to investments for rehabilitation of existing buildings and even to seismic mitigation campaigns of large building stocks. In order to evaluate the economic convenience of alternative mitigation strategies the expected seismic losses for different solutions (including the “do nothing”) in a given timeframe should be compared. The PEER performance-based earthquake engineering framework is normally adopted in order to compute losses. However, its application with a number of Non-linear Response History analyses as a basis for assessment of expected Engineering Demand Parameters and associated losses is an elaborate and time-consuming task, hardly employable for the assessment of earthquake losses for large building portfolios. This paper tests the applicability of a simplified pushover based approach for computation of earthquake losses. Referring to a case study building the results of application of classical PEER approach, i.e. based on Non-linear Response History analyses, are compared with the losses computed with pushover-based assessment, showing encouraging results. In addition, the simplified approach is used to investigate the effect of alternative retrofit strategies on the variation of expected seismic losses. 2805 Available online at www.eccomasproceedia.org Eccomas Proceedia COMPDYN (2017) 2805-2822