Evaluation of nonlinear structural analyses

Abstract

Nonlinear static procedures are now widely used in engineering practice to predict seismic demands in building structures under earthquake effects. Moreover, some seismic codes including new Turkish design code’06 have begun to consider them to aid in performance assessment of structural systems. The main aim of Performance assessment is to determine the performance target and performance criteria of structures. A critical component of evolving performance-assessment is the accurate estimation of seismic demand parameters. Nonlinear analyses are used in performance assessment. Time History analysis is the most reliable analysis method among the all nonlinear analysis methodologies. However Pushover analysis is become important due to its easy application comparing to time history analysis. It has been employed as a capacity analysis tool for the main purpose of seismic demand estimation. It is possible to mention three different pushover methodologies as classical, adaptive and energy based. With the increase in the number of alternative pushover procedures proposed in recent years, it is useful to identify the potential limitations of these methods and compare and contrast their effectiveness in simulating seismic demands at the structure, story and component level. The main purpose of this study is to evaluate the push over methodologies for frame type structures with different natural periods by comparing to nonlinear dynamic time history analyses. Each building is subjected to 30 near fault ground motions having different characteristics. 4-6-8-12-stroy frame type structures are selected and dimensioned regarding with the Turkish Design Code. These four structures are typical reinforced concrete (R/C) frame systems. First, classical pushover analysis is applied on the sample frame type structures, then, adaptive and energy based push over analyses are performed. To evaluate the results from the pushover analyses, nonlinear dynamic time history analyses are performed with earthquake ground motions. For the analyses, DRAIN 2DX program is conducted. Different types of loading patterns for classical pushover analyses are considered in the study. These are triangular, rectangular, and parabolic patterns taken from IBC. The lateral load patterns are applied step by step on the structures to sketch the pushover curves. Plastic hinge hypothesis is used to follow the structural behavior. The classical pushover analysis is based on the lateral load-deformation relationship in the aspect of material and geometric nonlinearity. The adaptive pushover analysis is based on a methodology changing with the inertia distribution for each modal shape and redistribution of loading. In this methodology it is possible to consider the higher mode effects and redistribution of inertia forces. There are so many different applications on adaptive pushover methodology. Chopra and Goel (Modal Pushover Analysis), Antoniou and Pinho, Mwafy and Elnashai, Gupta and Kunnath and Aydinoglu (Incremental Response Spectrum Analysis) methodologies are most important ones. Most of them are based on consideration of the higher mode effects. For the classical and adaptive pushover analyses, energy parameters are not taken into account. Structural damage is stated with the function of the structural displacement. However it is necessary to involve the energy parameters into pushover analyses for more realistic results. Energy based pushover analyses is applied by Montesfirst. Its aim is to consider energy parameters into pushover analyses. The structural displacements are calculated with the work from the pushover analyses. After conducting three different types of pushover methodologies, finally, time history analysis is realized. The displacements are found directly as a result of time history analysis which is not possible with any of pushover analyses. Dynamic loads are applied on the structures directly to get the structural displacement. For more realistic earthquake behavior estimation, more realistic nonlinear analysis is necessary. Therefore, to improve the pushover analyses for more realistic results is the primary focus of the earthquake engineering in the last decade. The importance of the pushover analysis explains the rising interest in recent developments of various methodologies. In this present study, the results of the pushover methodologies are evaluated with the time history analysis. The time history analysis is known as the most reliable nonlinear analysis since earthquake loads are applied on the structures directly in the analysis. As a result of this study, it is observed that, pushover analyses are to be developed for getting more approximate results to time history analysis. Over the three methodologies in pushover analyses, the best fitting result are observed in energy based pushover analysis.

Özet

Binaların deprem yükleri altındaki doğrusal olmayan davranışlarının belirlenmesinde birçok analiz yönteminden söz etmek mümkündür. Bunlara, doğrusal olmayan artımsal itme ve doğrusal olmayan zaman tanım alanında dinamik analiz yöntemleri örnek olarak verilebilir. Doğrusal olmayan zaman tanım alanında dinamik analiz, sismik davranışı belirlemede en iyi yöntem olarak kabul edilmektedir. Bununla birlikte doğrusal olmayan artımsal itme analizleri de kolaylıkları sebebiyle uygulamada yaygın olarak kullanılmaktadır. Artımsal itme analizleri, klasik, uyarlanmış ve enerji esaslı olmak üzere üç ana başlık altında toplanabilir. Bu çalışmanın amacı, binaların performansının belirlenmesinde kullanılan doğrusal olmayan artımsal itme analizlerinin, doğrusal olmayan zaman tanım alanında dinamik analizlerle karşılaştırmalı olarak değerlendirilmesidir. Bu amaçla çalışmada periyotları farklı, üç açıklıklı, 4, 6, 8 ve 12 katlı dört betonarme çerçeve tipi bina ele alınmıştır. Bu çerçeve tipi binalar için klasik artımsal itme analizleri, dikdörtgen, üçgen (IBC, k=1) ve parabol (IBC, k=2) yük etkileri altında gerçekleştirilmiştir. Ardından uyarlanmış ve enerji esaslı doğrusal olmayan artımsal itme analizleri gerçekleştirilmiştir. Elde edilen artımsal itme eğrileri, farklı 30 deprem verisi ile yapılan doğrusal olmayan zaman tanım alanında dinamik analiz sonuçları ile karşılaştırılmıştır.