Mevcut Betonarme Binaların Pera ( Hızlı Performans Değerlendirme Yöntemi ) İle Performans Analizinin Yapılması

Abstract

Son yıkıcı depremler, gelişmekte olan ülkelerde özellikle birçok mevcut binanın depreme karşı güvenli olmadığını göstermiştir. Bu binaların yıkılmasıyla oluşacak can ve mal kayıplarını azaltmak için gerekli değerlendirmelerin acilen yapılmasına ihtiyaç vardır. Kod tabanlı sismik değerlendirme yöntemleri, genel olarak ayrıntılı ve karmaşık yapısal analiz gerektirir. Bunun için gerekli maliyeti ve değerlendirme süresini azaltmak için basitleştirilmiş doğru değerlendirme yöntemlerinin gerekliliği ortaya çıkmıştır. Bu çalışmada, betonarme binalar için Hızlı Performans Değerlendirme Yöntemi (PERA) önerilen yöntem Deprem Bölgelerinde Yapılacak Binalar Hakkındaki Yönetmelik ve Riskli Binaların Tespit Edilmesine İlişkin Esaslar ile karşılaştırılmaktadır. Bu yöntem, betonarme binaların kritik deprem yükü altında olduğunu varsayar. Bu yönteme göre, kolonların elastik iç kuvvetleri, kesme ve eğilme kapasiteleri; bina ölçüleri,kolon boyut ve konumları, boyuna donatı oranı, etriye aralığı ve beton sınıfına uygun olarak elde edilir. Yapı genel performansı, tek tek kolonların etki/kapasite oranları yanı sıra en zayıf modları (akma/kopma) , eğilme ve kesme gerilme seviyelerine bağlı olarak belirlenir. Yanal ötelenmeler, basitleştirilmiş yaklaşımla hesaplanan kritik deprem yükünün yapısal performansın belirlenmesinde dikkate alınır. Güçlü ve zayıf kirişler, güçlü kolon-zayıf kiriş yada güçlü kiriş-zayıf kolon koşulları dikkate alınarak belirlenen kolon kesme kapasiteleri yapının doğal periyodunun belirlenmesinde dikkate alınır. Bu yöntemle elde edilen sonuçlar, Türkiye’deki tipik çerçeve sistemli betonarme yapıların 108 farklı durum için ayrıntılı analiz sonuçları ile karşılaştırılır. Öngörülen algoritma ve ayrıntılı yapısal performans değerlendirmesi arasında uyum olduğu elde edilir. Son olarak, önerilen yaklaşım öngörüleri, son iki yıkıcı depremden sonra Türkiye’de mevcut 9 yapıda gözlenen gerçek hasarlar ile karşılaştırılır. Bu karşılaştırmalarda kabul edilir düzeyde doğruluk görülmüştür. 17 Ağustos 1999’da Türkiye’nin Kocaeli İli’nde Rihter Ölçeğine göre büyüklüğü 7.4 olan yıkıcı bir deprem meydana geldi. Bu depremde bir çok bina ya hasar gördü yada kısmen veya tamamen çöktü. Bu çalışmada yıkımların en çok yaşandığı yerlerden biri olan Kocaeli İli Başiskele İlçesi’nde yer alan, 1999 Kocaeli Depremi’nden önce inşa edilen 6 bina ve depremden sonra inşa edilen 3 bina ayrıntılı olarak incelenmiştir. Seçilen 9 bina da 3 katlı olup zemin özellikleri bölgenin çoğunluğu Z3, sahil kesimleri ise Z4 grubundadır. Seçilen bu binalarda genel olarak düzenli olup betonarme çerçeve sistemden oluşmuştur. Bu 9 binanın performans değerlendirmesi PERA Yöntemi ile değerlendirilmiştir ve detaylı performans analizlerinin sonuçları ile karşılaştırılmıştır. Bu karşılaştırmalar incelenen betonarme binalar için PERA Yöntemi algoritmasının DBYBHYile uyumlu olduğunu göstermektedir.

Recent earthquakes, such as Kocaeli-Turkey (1999), Gujarat-India (2001), Bam-Iran (2003), Sumatra-Indonesia (2004), Kashmir-Pakistan (2005), Sichuan-China (2008), Haiti (2010), Tohoku-Japan (2011) and Van-Turkey (2011), which caused large number of casualties and injuries due to structural damages and collapses, have shown that a significant portion of existing buildings are not sufficiently safe against earthquakes. Therefore, the seismic safety of a vast number of existing buildings should be urgently evaluated for determining the vulnerable ones. Several seismic safety assessment procedures, ranging from street surveys to detailed vulnerability analysis, exist in the literature. Each of these methods requires procedures that demand various input parameters at different detail levels. The simplest seismic safety assessment procedure group consists of sidewalk (or street) surveys. FEMA 154 [1 and 2] and Sucuoglu et al. [3] are examples of these first-level approaches, which target to quantify and rank buildings that are seismically hazardous, before a detailed assessment is carried out. The FEMA 154 Rapid Visual Screening procedure [1 and 2] was developed for twelve different types of structural systems. The Basic Structural Hazard Score, determined based on the type of the structural system, is modified via score modifiers, which are related to the observed performance attributes such as visual condition, number of stories, vertical and plan irregularities, comparison of the design and construction dates and the soil type. Final scores typically range from zero to seven, with higher score corresponding to better seismic performance expected for the building. The method introduced by Sucuoglu et al. [3] is applicable to low- and mid-rise reinforced concrete buildings up to six stories. This sidewalk survey aims to obtain a performance score for each of the buildings in the investigated region, so that they can be ranked with respect to their seismic risk levels. Similar to the FEMA 154 method [1 and 2], this method also modifies the basic score of the building with vulnerability score-modifiers. The basic score depends on type of the structural system, seismicity and local site conditions of the building, defined in terms of peak ground velocity. The vulnerability scores depend on structural attributes such as presence of soft story, heavy overhangs, short columns, pounding potential, as well as apparent visual quality and topographical effects. Vulnerability parameters ranging between zero and one are used to modify the vulnerability scores, which in turn are subtracted from the basic score. Methodologies including Japanese Seismic Index Method [4], Hassan and Sozen [5], Yakut [6], P25 Method [7], and NZSEE Method [8] can be pronounced among more detailed preliminary assessment approaches. The Japanese Seismic Index Method [4] consists of three different levels of screening and/or assessment procedures. The first level is the simplest and the most conservative approach. Only the compressive strength of concrete and the cross-sectional areas of columns and walls are considered for estimating the seismic capacity of the building, while ductility characteristics are neglected. In the more detailed second and third levels, ultimate lateral load capacities of the frames and shear walls are evaluated using material and cross-sectional properties, together with reinforcement details which require in-situ structural drawings. The second level procedure evaluates the seismic capacity of the building with the strong beam – weak column assumption, so that the strength and ductility of only the vertical members are considered. The third level procedure considers the strength of beams in addition to the strength of columns and walls, for evaluation of the seismic capacity. It should be noted that a number of studies that aim to adopt the Japanese Seismic Index Method [4] to Turkish buildings is available in the literature (Baysan, [9], Ilki et al., [10], Boduroglu et al., [11], Boduroglu et al., [12], Ozdemir et al., [13]). The Hassan and Sozen [5] method follows the approach introduced by Shiga et al. [14] after the Tokachi-Oki earthquake of 1968. This method is applicable to low- to mid-rise reinforced concrete buildings and considers only the cross-section dimensions and orientations of the vertical members, such as columns, shear walls, and infill walls. In this method, the total column area at the base of the building is divided by the total floor area of the building for computing the column index (CI). Similarly, the shear wall and infill wall areas in one direction are divided by the total floor area above the base so that the wall index (WI) for that direction is obtained. Finally, the column and wall indexes are graphically evaluated. Accordingly, as the wall and column indexes of the building become smaller, the vulnerability of the building increases. The procedure proposed by Yakut is recommended for low- to mid-rise reinforced concrete frame buildings with and without shear walls. The method estimates the elastic base shear capacity of the building using the dimensions, orientation and concrete strength of the structural components at the ground floor of the building. The contribution of the infill walls are also considered for calculation of the Basic Capacity Index, which is the ratio of the estimated yield base shear of the building with infill walls to the code required base shear. Then the Basic Capacity Index is modified such that the effects of the construction quality and architectural features (such as vertical and plan irregularities) are also reflected in the estimation. Finally, the obtained Capacity Index is compared with a cutoff value for reaching a decision on the vulnerability of the building. The P25 Scoring Method [7], which aims to identify collapse-vulnerable structures, was developed by using a database of 323 buildings, which have experienced varying levels of damage during previous earthquakes in Turkey. The method is based on seven different scores for corresponding failure modes and their interactions, as a function of their estimated relative importance. The method considers several parameters such as concrete quality, seismicity, pounding, potential short column, corrosion, irregularities in plan and elevation, confinement, foundation type, foundation depth, ground conditions, heavy overhangs and heavy façade elements. The NZSEE Initial Evaluation Procedure (IEP) [8] involves making an initial assessment of the performance of existing buildings against the standard required for a new building (i.e. “percentage new building standard” (% NBS)). Accordingly, the nominal % NBS value is determined by considering the date of design, seismic zone and soil type together with the estimated period of the building calculated through simple equations given for different structural system types. The nominal % NBS value is then multiplied by near fault, hazard, return period, ductility, and structural performance scaling factors, so that the baseline % NBS is obtained. In the next step, the baseline % NBS is modified by the Performance Achievement Ratio (PAR) which covers critical structural weaknesses such as plan and vertical irregularities, short columns, pounding potential, site characteristics, and other factors that can be included by the engineers. Finally, the assessment of the building is completed considering the resulting % NBS values for longitudinal and transverse directions. The building is classified as potentially earthquake prone for % NBS values less than 33, and a more detailed evaluation is required. Insignificant earthquake risk is foreseen for buildings with % NBS values greater or equal to 67. For 33< % NBS <67 a more detailed evaluation is recommended. While these methods are useful and valuable tools for rapid seismic safety assessment, they present several drawbacks, which necessitate more accurate yet simple methods developed based on structural mechanics principles and capable of considering different potential failure modes of structural members. Main disadvantage of these methods stem from uncertainties they include in terms of risk scores and threshold performance values since they are generally based on expert judgment or are calibrated considering statistical data representing a certain earthquake ground motion, ground condition and structural typology. Several other seismic assessment methodologies have also been proposed based on inelastic displacement demand and/or considering probabilistic approaches such as Ruiz-Garcia and Miranda [15], Priestley [16], Chandler and Mendis [17], Jeong et al. [18], and Iervolino et al. [19], whereas Lupoi et al. [20] and Kalkan and Kunnath [21] have compared detailed linear and nonlinear static assessment methodologies in their studies. In this study, a performance based rapid seismic safety evaluation (PERA) methodology is proposed for reinforced concrete frame structures, for which the effect of first vibration mode is dominant in the seismic response. The proposed methodology makes use of several approaches of Muto [22], member tributary area concept, and other certain simplifications and assumptions related to structural analysis and performance based assessment. For the estimation of member damages and overall structural seismic performance evaluation, performance criteria of the Turkish Seismic Design Code (TSDC) [23] are taken into account. During seismic performance evaluation, the axial-flexural and shear capacities of all vertical structural members, considering the actual type of longitudinal and transverse reinforcing bars, diameter and spacing of transverse bars, and estimated concrete quality are taken into account, together with certain assumptions related to the geometric ratio and configuration of the vertical bars in the columns. In addition, structural irregularities as defined by the TSDC [23] are also considered during evaluation. Consequently, while the amount of data required is not remarkably more than the rapid and preliminary assessment methods outlined above, determination of the type of reinforcing bars, stirrup spacing, and concrete quality (with limited number of tests), together with proper consideration of different failure modes, make the proposed algorithm significantly more realistic compared with existing methodologies. More importantly, since the seismic safety evaluation is conducted considering the provisions of the TSDC [23], potential problems that other rapid assessment methodologies can create, due to non-compliance with the existing code, are minimized. Recent destructive earthquakes have shown that many existing buildings, particularly in developing countries, are not safe against seismic actions. To mitigate the collapse of these buildings, and reduce casualties and economic losses, assessment and rehabilitation actions are urgently needed. Since code-based seismic safety evaluation methods generally require detailed and complex structural analysis, the necessity for simplified, yet sufficiently accurate evaluation methods emerges for reducing cost and duration of assessment procedures. In this study, a performance based rapid seismic safety assessment method (PERA) is proposed for reinforced concrete buildings. The proposed method assumes that ground story of the building is critical against seismic loads. According to this method, elastic internal force demands of columns of ground story and their shear and axial-flexural capacities are obtained in accordance with the as-built structural drawings, the actual type and estimated ratio of longitudinal bars and the actual type and spacing of transverse bars and concrete quality. The overall structural performance is determined based on the demand/capacity ratios of individual columns, as well as their failure modes (brittle/ductile), confinement characteristics, and levels of axial and shear stresses. The lateral drift of the critical story, calculated through a simplified approach, is also taken into account during determination of the global structural performance. It should be noted that the strength and stiffness of the beams are taken into account during estimation of natural period of the building, determination of column shear demands considering strong column – weak beam or strong beam – weak column conditions, and estimation of contra flexure points on the columns. The predictions of this method are compared with the results of conventional detailed seismic safety assessment analyses carried out for 108 different cases representing typical reinforced concrete frame buildings in Turkey. Good agreement is obtained between the predictions of the proposed algorithm and code-based structural performance assessment procedures. Finally, predictions of the proposed approach are compared with actual damages observed in 9 existing buildings in Turkey after destructive earthquakes that have occurred during the last two decades. These comparisons also point to an acceptable level of accuracy and sufficient conservatism for the methodology proposed.