THESES ABSTRACTS

Home

Türkçe

A NEW METHODOLOGY FOR SEISMIC VULNERABILITY ASSESSMENT OF
EXISTING BUILDINGS IN TURKEY

PAY, Ali Cihan
M.S. Thesis
Supervisor : Prof. Dr. Güney ÖZCEBE
Co-supervisor :-
August 2001,
175 pages

In this study, a new methodology is presented to predict the seismic vulnerability of reinforced concrete structures by statistical analysis based on a number of structural parameters selected on the basis of engineering judgment and observations. The available data collected after the 17 August and 12 November 1999 earthquakes in Bolu, Düzce, and Kaynasli are examined by utilizing “discriminant analysis”. Considering the characteristics of the damaged structures, the following parameters are decided as the basic parameters for the assessment of vulnerability: number of stories, degree of overhang, soft story, redundancy, and square root of sum of squared moment of inertias (SRSSI). However, during the discriminant analysis, two parameters, overhang and soft story, are found to be statistically insignificant. Therefore, the statistical analysis is based on the remaining three parameters; namely: number of stories, redundancy, and SRSSI. The proposed method is checked with respect to the available data collected after the 1992 Erzincan earthquake and performed quite satisfactorily..

Keywords : reinforced concrete, seismic vulnerability, discriminant analysis, earthquakes

STOCHASTIC METHODS FOR THE ESTIMATION OF POTENTIAL SEISMIC DAMAGE

ASKAN, Aysegul
M.S. Thesis
Supervisor : Prof. Dr. Semih YUCEMEN
Co-supervisor :-
July 2002,

Due to uncertainties involved in both occurrences of earthquakes and structural responses, earthquake damage prediction has to be treated in a probabilistic manner. In this study, three stochastic methods, for the prediction of potential seismic damage to low and mid-rise reinforced concrete buildings in Turkey, are presented. A damage probability matrix (DPM) expresses what will happen to buildings, designed according to some particular set of requirements, during earthquakes of various intensities. In this study, as the first approach, “best estimate” damage probability matrices for each seismic zone are developed by combining expert opinion and the damage statistics compiled from the recent earthquakes occurred in Turkey. Second approach involves a reliability-based model, which treats the earthquake force and seismic resistance as random variables. This model expresses potential seismic damage in the form of a damage rate distribution, which is a function of modified Mercalli intensity or peak ground acceleration. As the third methodology, discriminant analysis technique is utilized to carry out a statistical analysis on the damage data compiled during recent earthquakes that occurred in Turkey. Based on the classification procedure involved in this technique, the damage state probabilities corresponding to modified Mercalli intensity levels of the relevant earthquakes are obtained.

Keywords: Stochastic methods, structural response, damage prediction, seismic damage, earthquakes

SEISMIC RETROFIT OF BRICK INFILLED R/C FRAMES WITH LAP SPLICE PROBLEM IN COLUMNS

AKGUZEL, Umut
M.S. Thesis
Supervisor : Prof. Dr. Turan ÖZTURAN
Co-supervisor :-Assistant Prof. Dr. Şevket ÖZDEN
January 2003,

Recent earthquakes revealed that many existing structures located in seismically active regions of Turkey have inadequate lateral strength, stiffness or ductility. Lately, a significant amount of research has been devoted to the study of various strengthening techniques to enhance the seismic performance of the predominant structural system of the region, which is reinforced concrete frames with unreinforced masonry infills. In this context, an alternative strengthening method consists of externally applied carbon fiber reinforced polymers (CFRP) over the brick infilled reinforced concrete frames has been proposed and investigated. The use of CFRP materials offers important advantages such as ease of application, minimum disturbance to the occupants and savings in construction cost and time in addition to their advanced mechanical properties. Five specimens were tested to highlight the effect of brick infill and epoxy bonded CFRP overlays on the strength and behavior of poorly detailed reinforced concrete frames. The main deficiencies of the one-third scale one-bay, two-story frames tested were low concrete strength, insufficient column lap splice length, poor confinement, and inadequate anchorage length of beam bottom reinforcement. In all specimens beams were stronger than columns and no joint shear reinforcement was used. Strength, stiffness, interstory drift, energy dissipation capacity, and viscous damping ratio of the specimens were examined by evaluating the test results. Recommendations are also presented on the usefullness of the proposed strengthening technique.

Keywords: seismic strengthening, CFRP, brick infilled reinforced concrete frames, cyclic loading.

CARBON FIBER REINFORCED MASONRY INFILLED
REINFORCED CONCRETE FRAME BEHAVIOUR

MERTOL, Halit Cenan
MS Thesis
Supervisor: Prof. Dr. Tugrul TANKUT
Co-supervisor: -
June 2002, 67 pages

Seismic strengthening is a very important problem in Turkey, because there are a lot of buildings which do not satisfy the code requirements. So a strengthening method must be found for this kind of buildings. And it must be an easily implemented method with minimum time consuming. But the most important subject in this method is to give the minimum disturbance to the user as the strengthening actions continue. In other words, the building must function properly as strengthening takes place. On the other hand, in Turkey, since a lot of buildings have brick masonry infill walls, the combination of giving minimum disturbance and brick infill walls reveals the aim of this research. In the tests carbon fiber reinforced polymer strengthening was used to achieve this aim. The implementation of this material is very easy and the strength is very high. So the combination of carbon fiber sheets and the masonry infill walls were used in this thesis. A behavior like a shear wall behavior was aimed with this combination. For this purpose, two two-story, one-bay, 1/3 scale reinforced concrete frames infilled with plastered brick masonry were constructed. These frames had poor concrete quality. The frames did not have confinement of the stirrups at the column and beam ends. The frames were tested under the reversed cyclic loading. One frame was used as a reference frame (tested with no strengthening) and the other frame's only infill walls were strengthened over the plaster. At the end of the tests, strength, stiffness, energy dissipation and story drift characteristics of the specimens were examined. Since this research is only a small part of a greater research, the results are used as a basis for the following researches.

Keywords: seismic strengthening, reinforced concrete frames, brick masonry infill, CFRP, cyclic loading

BEHAVIOR OF BRICK INFILLED REINFORCED CONCRETE FRAMES STRENGTHENED BY CFRP REINFORCEMENT: PHASE I

KESKIN, Riza Secer Orkun
MS, Civil Engineering
Supervisor: Prof. Dr. Güney Özcebe
Co-supervisor:
July 2002, 140 pages

In Turkey, there exist a great number of structures that do not fulfill the requirements of the Turkish Seismic Code. A new, rapid and easy strengthening method that would not disturb the serviceability of the structures should be developed in order to prevent collapse of those structures. This study was a part of the research program that was developed in METU-Structural Mechanics Laboratory for this purpose. The aim of the research program was to study the behavior of brick infilled reinforced concrete frames strengthened by carbon fiber reinforced polymer (CFRP) reinforcement. In this part of the research program, two one-third scale, one-bay, two-story reinforced concrete frames were infilled with hollow clay tiles and strengthened with CFRP. The main test variable was detailing of CFRP reinforcement. Test specimens were tested under reversed cyclic loading. Axial loads were kept constant throughout the tests. One of the test specimens failed prematurely, whereas the other one performed well. Test results were evaluated in terms of strength, stiffness, energy dissipation and interstory drift characteristics. In addition to experimental study, an analytical study was performed on DBI Building in Dinar. The building was moderately damaged after the Dinar earthquake in 1995, and, then, rehabilitated with reinforced concrete infill walls. Analyses on four different arrangements of CFRP reinforcement were performed and compared with the analysis of the building rehabilitated with reinforced concrete infill walls. As a result, a re-design criterion was suggested for pre-earthquake rehabilitation of structures.

Keywords: strengthening, CFRP, brick infilled reinforced concrete frames, reversed cyclic loading

BEHAVIOR OF BRICK INFILLED REINFORCED CONCRETE FRAMES STRENGTHENED BY CFRP REINFORCEMENT: PHASE II

ERDURAN, Emrah
MS
Supervisor: Prof. Dr. Güney Özcebe
Co-supervisor: -
July 2002, 77 pages

Seismic pre-rehabilitation of poorly designed and/or constructed structures is a challenging engineering problem as far as huge stock of buildings is concerned. This problem should be solved in a way that the residents of the buildings should not be enforced to leave the structure during the rehabilitation period. Thus, the method of strengthening must be effective and practical to apply. Carbon Reinforced Polymer (CFRP) seems to be a solution for this problem owing to its high strength and particularly ease of application. In this study, CFRP is applied on mainly the existing plastered brick infill walls of brick infilled reinforced concrete frames. For this, two one-bay, two-story, one to third scale specimens were constructed and tested under reversed cyclic loading. The specimens were constructed with the most common deficiencies observed in practice. The test results were evaluated in terms of strength, stiffness, interstory drift, and energy dissipation capacity characteristics. A model for composite material was derived using the test results. This model was used to develop design criteria for strengthening of structures using CFRP.

Keywords: seismic strengthening, CFRP, brick infilled reinforced concrete frames, cyclic loading.

SEISMIC STRENGTHENING OF REINFORCED CONCRETE FRAMES WITH PRECAST CONCRETE PANELS

DUVARCI, Murat
MS
Supervisor: Prof. Dr. Tugrul TANKUT
Co-supervisor: -
April 2003, 90 pages

The importance of seismic rehabilitation of structures had been better understood after the severe earthquakes occurred in Turkey. A large number of reinforced concrete buildings damaged in these earthquakes needed repair and/or strengthening. In Turkey, the major problems of buildings are inadequate stiffness and ductility. To overcome these problems, new easy and rapid strengthening methods are being developed in METU Structural Mechanics Laboratory.

The aim of the method reported in this thesis is to strengthen the buildings by using precast concrete panels. This is an easy method which results in a rapid construction and least disturbance given to the occupants. In the study, two preliminary test were conducted to verify the proper functioning of the newly developed test setup and then three brick infilled, one-third scale, one bay, two story reinforced concrete frames which reflect the common deficiencies of the buildings in Turkey were constructed as test specimens. First a reference specimen was tested and the other two specimens were strengthened by using precast concrete panels. Test specimens were tested under reversed cyclic loading and axial load which was kept constant throughout the tests.

Test results were evaluated in terms of strength, stiffness, energy dissipation and interstory drift characteristics.

Keywords: seismic rehabilitation, precast concrete panels, brick infilled, reinforced concrete frames, reversed cyclic loading.

SEISMIC STRENGTHENING OF R/C FRAMED STRUCTURES
BY PRECAST CONCRETE PANEL INFILLS

BARAN, Mehmet
Ph.D
Supervisor: Prof. Dr. Tugrul TANKUT
Co-supervisor: -
Ongoing

In Turkey, buildings suffer following problems: insufficient lateral stiffness, non-ductile members, bad detailing and low concrete quality. Even if there is no any strength problem, insufficient lateral stiffness can cause significant non-structural and structural damage during severe earthquakes.

In the solution of this problem, infilled frame approach can be applied by using cast-in-place reinforced concrete, masonry (or strengthened masonry), diagonal bracing and precast panel.

System behavior improvement by introducing shear walls and R/C infills becomes a feasible solution when frames have soft story, inadequate lateral stiffness, undesirable hinging mechanism (weak column-strong beam) and the number of structural member which need to be strengthened is high. These are the problems from which most of the buildings in Turkey suffer from. However, strengthening of R/C framed structure by using cast-in-place R/C infills leads to a huge construction work. Hence, using prefabric panel infills or carbon fibers for infills become attractive from the point of view of:
i) minimal disturbance to the occupant,
ii) construction work mass.

The aim of this experimental study is to observe the seismic behavior of strengthened R/C frames by precast concrete panel infills. This study also involves:
i) to modify the existing test set-up for one-bay, two-story frames,
ii) to design different types of precast concrete panel infills,
iii) to design and investigate the seismic performance of various connection types between panel to panel and panel to frame.

Keywords: earthquake, seismic risk, structural rehabilitation, inadequate lateral strength, inadequate lateral stiffness, precast panel infil.

STRENGTHENING OF BRICK INFILLED RC FRAMES WITH CFRP REINFORCEMENT - GENERAL PRINCIPLES

AKIN, Emre
Ph.D
Supervisor: Prof. Dr. Güney ÖZCEBE
Co-supervisor: -
Ongoing

With the developing technology, composite materials become an alternative solution for seismic strengthening of structures. However, there are not adequate studies about their usage in system strengthening. If they can be implemented in system strengthening of reinforced concrete frames efficiently, the rehabilitation of poorly designed and/or constructed structures may be completed easily in relatively short time. This study is a part of a large project initiated at METU, on strengthening of the existing reinforced concrete structures.
The main objective of this study is to increase the seismic performance of poorly designed and constructed reinforced concrete frames. Preventing the collapse of the structures during a major earthquake is determined as the level of the aimed seismic safety. Strengthening of structures using CFRP is a part of this project. The objective of the experimental part of this study is to develop design details for the strengthening of brick infilled RC frames by using CFRP reinforcement. In the analytical part of the study, it is aimed to develop design criteria for strengthening of structures using CFRP.

Keywords: seismic strengthening, reinforced concrete frames, brick masonry infill, CFRP Strengthening, design guidelines, aspect ratio, wall ratio, cyclic loading ,

IMPROVING DUCTILITY AND SHEAR CAPACITY OF REINFORCED CONCRETE COLUMNS
WITH EXTERNAL PRESETRESSING

ÖZCAN, Okan
Ph.D
Supervisor: Prof. Dr. Güney ÖZCEBE
Co-supervisor: -

The performance of reinforced concrete columns during recent earthquakes has clearly demonstrated the possible failures associated with inadequate transverse reinforcement. The transverse reinforcement requirements of older codes were less stringent than those of present standards. Hence, many of the reinforced concrete columns built prior to the early 1970s could experience the same failures as those seen in recent earthquakes. Many studies were done by applying different retrofitting techniques for reinforced concrete columns that have inadequate transverse reinforcement. A new retrofitting technique was developed and tested at the University of Ottawa by Saatçioglu et al. This technique, the Retro-Belt retrofitting system, consists of prestressing strands and specially designed anchors placed and stressed externally on the column as individual hoops. The effectiveness of Retro-Belt system was shown for Canadian applications. However, the Turkish construction practice is considerably different than the Canadian practice. In Turkey, the frame members, i.e. columns and beams, are considerably deficient form the seismic detailing point of view. In addition, the poor material qualities further worsen the seismic safety for the frames built in Turkey. Therefore, in order to be able to use the Retro-Belt effectively in Turkey a series of experimental work need to be done. In this experimental work necessary calibrations to adjust the level of pre-stressing, hoop spacing and strand size form the parameters of design, and can be selected to suit the size and retrofit needs of the column in our country.

Keywords: reinforced concrete, column, improving shear capacity, improving ductility, external prestressing

DEVELOPMENT OF DAMAGE FUNCTIONS FOR REINFORCED CONCRETE COMPONENTS

ERDURAN, Emrah
Ph.D
Supervisor: Assistant Prof. Dr. Ahmet YAKUT
Co-supervisor: -

Recent focus on developing procedures for predicting earthquake imposed damage on buildings has found broad appeal because there has been extensive need for such procedures. These procedures are used for the purpose of evaluating performance of existing buildings as well as assessing building vulnerability. Predicting vulnerability of a whole structure is not easy to determine due to lack of proper experimental and observed data. For this reason, the trend has moved towards evaluating the whole structure at the level of its components. In this framework, research has been undertaken to develop damage curves for reinforced concrete members. Columns, beams and shear walls are considered as primary components of reinforced concrete buildings and infill walls as secondary members. Since available experimental studies on this subject are limited, a broad range of parameters that influence the damageability of reinforced concrete members are investigated analytically. Analytical results are calibrated with available experimental data to determine the significant parameters that need to be considered in the damage curves. Damage curves expressed in terms of inter story drift ratio will be developed for various levels of ductility.

Once component level investigations are completed, story and then building level damage functions will be developed. For a given story drift obtained from nonlinear analysis, component damage functions will be combined using weighting coefficients that reflect the importance of each component. The component importance factors depend on the role of the member in resisting seismic forces. These factors will be derived from nonlinear analyses of some typical frames, including various failure mechanisms. The story level damages will then be combined based on the relative importance and location of the floors to obtain building damage. This damage will be expressed in terms of spectral displacement computed from the top floor displacement.

The building damage function presents a plot of expected damage for a given level of ground motion represented by the spectral displacement.

Keywords:

PROBABILISTIC SEISMIC LOSS ESTIMATION METHODOLOGY FOR TURKEY

ÖZTÜRK, Yilmaz Nazan
Ph.D
Supervisor: Prof. Dr. Semih YÜCEMEN
Co-supervisor: Prof. Dr. Güney ÖZCEBE

In this study, it is aimed to develop a probabilistic seismic loss estimation methodology for Turkey. Basic steps to be followed are seismic hazard analysis, vulnerability analysis and loss estimation.

The probabilistic hazard analysis will be performed to obtain maximum ground motion parameters (e.g.. peak ground acceleration, spectral acceleration) over the site with 10 percent probability of being exceeded in 50 years. Seismic hazard analysis will involve compilation, preparation and analysis of earthquake catalog data, earthquake source modeling, and attenuation relationship and site properties. Available earthquake catalog will be updated by considering the recent earthquakes that occurred in Turkey. A local attenuation relationship will be derived based on ground motion records obtained from the earthquakes in Turkey. A computer program will be utilized to perform the seismic hazard analysis.

Vulnerability analysis will be performed to obtain building fragility curves as a function of ground motion. Fragility curves will provide estimates of the cumulative probabilities of being in, or exceeding, slight, moderate, extensive and complete damage for the given level of ground shaking. To obtain these curves, statistical methodologies will be applied to seismic damage data compiled from past earthquakes. In this study, mainly, low to mid-rise reinforced concrete frame structures with or without shear walls and masonry structures will be considered.
In the loss estimation part, unit construction costs of the building stock will be obtained. Then, for each state of physical damage, respective repair costs will be calculated and normalized with respect to the replacement value. Then, a relationship between the predicted ground motion at the site and the expected loss will be established for each construction type.

The computations involved in the above steps will be conducted by preparing a software program and its initial implementation will be done for a specific site.

Keywords:

SEISMIC VULNERABILITY ASSESSMENT OF EXISTING REINFORCED CONCRETE BUILDINGS IN TURKEY

AYDOGAN, Volkan
MS
Supervisor: Prof. Dr. Güney ÖZCEBE
Co-supervisor: -

A statistical procedure, called discriminant analysis, is used to develop a model for the preliminary assessment of the seismic vulnerability of reinforced concrete buildings. The earthquake damage data compiled in Duzce province after the 17 August 1999 Kocaeli and 12 November 1999 Duzce earthquakes formed the damage database. Number of stories, minimum normalized lateral stiffness index, minimum normalized lateral strength index, normalized redundancy score, soft story index and overhang ratio are selected as the basic damage inducing parameters. Two discriminant functions are derived in terms of these variables considering immediate occupancy and life safety performance levels.

In the proposed preliminary seismic vulnerability assessment model, the discriminant scores obtained from these two discriminant functions are combined in an optimal way and is used to classify existing buildings as "safe", "unsafe" and "requires further evaluation". The optimality criterion imposed into the model is the minimization of the misclassification rate of damage states causing collapse. The validity of the proposed model is checked by using the seismic damage data associated with recent earthquakes that occurred in Turkey. The consistency between observed damage and the predictions of the proposed model supports the effectiveness of the proposed model.

Keywords: seismic vulnerability, discriminant analysis, earthquake damage estimation, reinforced concrete, Duzce earthquake

SEISMIC STRENGTHENING OF R/C FRAMED STRUCTURES BY PRECAST CONCRETE PANEL INFILLS

SÜSOY, Melih
MS
Supervisor: Prof. Dr. Tugrul TANKUT
Co-supervisor: -

Over 90 percent of the land area of Turkey lies over one of the most active seismic zones in the world. Hazardous earthquakes frequently occur and cause heavy damage to the economy of the country as well as human lives. Therefore, seismic research is considered as a crucial effort. It is not possible to predict the time, location or size of an earthquake, however, it is possible and necessary to design and construct buildings capable of withstanding earthquakes of considerable size.

Unfortunately, the majority of buildings in Turkey do not have enough seismic resistance capacity. The most commonly observed problems are faulty system configuration, insufficient lateral stiffness, improper detailing, poor material quality and mistakes during construction. Due to this fact, extensive damage is experienced after every major earthquake. In our country, starting with 1992 Erzincan earthquake, extensive structural rehabilitation have been done after the major earthquakes. Also, importance of rehabilitation before occurrence of an earthquake in order to improve the capacity to a desired level is appreciated. Repair or strengthening can be applied to individual structural members, or system improvement approach can be followed, which is generally provided by introducing shear walls and R/C infills. However, strengthening of R/C framed structure by using cast-in-place R/C infills leads to a huge construction work and is time-consuming. On the other hand, using prefabric panel infills can be preferred as a more feasible, rapid and easy technique during which the structure can remain operational.

The aim of this experimental study is to observe the seismic behavior of strengthened R/C frames by precast concrete panel infills by testing different types of panel and connection designs.

Keywords: earthquake, seismic risk, structural rehabilitation, lateral stiffness, shear wall, reinforced concrete infill, precast panel infill

PREPARATION OF AN INTERNET BASED 2D COMPUTER SOFTWARE FOR THE STRUCTURAL EVALUATION AND SEISMIC VULNERABILITY ASSESSMENT OF EXISTING R/C BUILDINGS

YALIM, Baris
MS
Supervisor: Assistant Prof. Dr. Ahmet TÜRER
Co-supervisor: -

Structural evaluation and seismic vulnerability assessment of Reinforced Concrete (R/C) buildings have especially become the focus of many researches in Turkey and abroad especially after the August 17, 1999 earthquake causing major life and property losses. A devastating earthquake being expected in Istanbul-Marmara region raises many questions on how well the existing buildings are constructed and whether they can stand a major earthquake. Evaluation of existing buildings for seismic vulnerability requires time consuming input preparation (pre-processing), modelling, and post processing of analysis results. The objective of the study is to perform automated seismic vulnerability assessment of existing R/C buildings automatically over the internet by asking internet users to enter their building related data, and streamlining the modelling-analysis-reporting phases by intelligent programming. The internet based assessment tool is prepared for two levels of complexity: (a) the detailed level targets to carry out seismic evaluation of the buildings using a linear structural analysis software developed for this study; (b) the simplified level produces seismic evaluation index for buildings, based on simple and easy to enter general building information which can be entered by any person capable of using an internet browser. Detailed level evaluation program includes a user friendly interface between the internet user and analysis software, which will enable data entry, database management, and online evaluation/reporting of R/C buildings. Building data entered by numerous users over the internet will also enable formation of an extensive database of buildings located all around Turkey.

36 buildings from Düzce damage database, generated by the cooperation of Scientific and Research Council of Turkey (TÜBİTAK) and Structural Engineering Research Unit (SERU) after the 17 August 1999 Kocaeli and the 12 November 1999 Düzce earthquakes, are used in the analyses to identify relationship between calculated indices and observed damage levels of buildings, which will enable prediction of building damage levels for future earthquakes. The research is funded by Science Research Program (BAP 2003-03-03-03), NATO-SfP 977231, and TUBITAK ICTAG-I574 projects.

The contribution of the research is composed of a) online building index -performance analysis/evaluation software which might be used by any average internet user, b) an ever-growing R/C building database entered by various internet users.

Keywords: Earthquake, seismic vulnerability, internet-based, software development, Düzce earthquake:

PREPARATION OF AN AUTOMATED 3D-FE MODEL GENERATOR COMPUTER SOFTWARE
FOR CONDITION EVALUATION OF EXISTING R/C BUILDINGS FOR EARTHQUAKE VULNERABILITY

ÖKSÜZ, Arif
MS
Supervisor: Assistant Prof. Dr. Ahmet TÜRER
Co-supervisor: -

The objective of this study is to develope of a general computer program that will streamline the model generation of three dimensional (3D) finite element models (FEM) for existing R/C buildings. The program will automatically generate a model, execute analysis using externally available software, and read the output files of the analysis for further earthquake vulnerability assessment process. The developed program will have a user friendly interface to input detailed structural information of a building from existing or field measured plans. The FEM generated by the program will have the flexibility to be further modified using other commercially available software and be used for push-over and/or nonlinear analysis. A number of buildings that were damaged during recent earthquakes will be analyzed using this developed tool. The vulnerability assessment results of case studies will be compared against the available building performances during recent earthquakes.

Keywords:

SEISMIC VULNERABILITY ASSESSMENT OF EXISTING BUILDINGS USING NEURAL NETWORKS

GÜLER, Altug
MS
Supervisor: Assistant Prof. Dr. Ahmet TÜRER
Co-supervisor: Assistant Prof. Dr. Ahmet Türer

The objective of this study to generate a Neural Network (NN) which will simulate the human brain to find out inherent relationships between the selected parameters of an existing R/C building and the seismic performance for a seismic excitation with certain properties. The existing information from past earthquakes and the performance of buildings experienced those earthquakes will be used to form the training database for the NN. The accuracy of the trained NN will be tested by a randomly selected subset of building performance database which were not used among the training set. A number of earthquake sites and building performances will be used to converge to a more global evaluation tool which will also include soil characteristics as a parameter. The finalized tool is expected to serve for fast evaluation of existing buildings for seismic performance and vulnerability assessment.

Keywords:

EXPECTED DAMAGE DISTRIBUTION IN ISTANBUL: CASE STUDIES

KÜÇÜKÇOBAN, Sezgin
MS
Supervisor: Assistant Prof. Dr. Ahmet YAKUT
Co-supervisor: -

The occurrence of two recent major earthquakes in Turkey prompted seismologist and geologists to conduct studies to predict magnitude and location of a potential earthquake that can cause substantial damage in Istanbul. Many scenarios are available about the extent and size of the earthquake. Moreover, studies have recommended rough estimates of risk areas throughout the city to trigger responsible authorities to take precautions to reduce the casualties and loss for the earthquake expected. Most of these studies, however, adopt available procedure by modifying them for the building stock peculiar to Turkey. The assumptions and modifications made are too crude and thus are believed to introduce significant deviations from the actual case. To minimize these errors and use specific damage functions and capacity curves that reflect the practice in Turkey, a study was undertaken to predict damage pattern and distribution in Istanbul for two scenario earthquakes proposed by JICA. The success of these studies strongly depends on the quality and validity of building inventory and site property data. The building stock in Istanbul will be broken down into 4 groups: Reinforced Concrete Frames with infill Walls, Reinforced Concrete Shear Wall, Prefabricated and Unreinforced Masonry Buildings. Within each group three height classes, low, mid and high, will be considered.

Building damage functions and capacity curves developed from the studies conducted in Middle East Technical University will be used. A number of proper attenuation relations and suitable amplification coefficients will be employed. The study will focus mainly on developing a GIS based software to carry out all computations and present results.

The results of this study will reveal a more reliable picture of the physical damage distribution expected in Istanbul.

Keywords:

A METHODOLOGY FOR DETERMINATION OF
PERFORMANCE BASED DESIGN PARAMETERS

YAZGAN, Ufuk
MS
Supervisor: Prof.Dr. Polat GULKAN
Co-Supervisor: Assistant Prof. Dr. Ahmet YAKUT

It is an accepted fact that both the structural and nonstructural earthquake damage results primarily by lateral displacements taking place during the earthquake. However, current building codes which use approximate lateral forces for the design of structural members give secondary importance to lateral displacements. Adequate displacement ductility of the structure is achieved by proper detailing of the structural members as given in the codes. The usual practice for the estimation of displacements is factoring of lateral displacements found from linear analysis with an amplification coefficient that is proposed for taking into account the inelastic deformation that may occur in the structure during earthquake. Then, as a final step, these amplified displacements are checked for serviceability limits by using with the allowable upper limits given in the codes. Although, the primary cause of damage resulting from near field earthquakes are due to excessive lateral displacements, current building codes lack explicit consideration of displacement demands.

Based on these observations, recently there has been a growing interest on displacement based design methods. Unlike force-based methods, the performance-based design methods use target lateral displacements to fulfil the required design objectives. The target displacements can be either the maximum roof displacement (global deformation capacity) or the maximum inter-story drift (maximum lateral deformation between two consecutive stories). The inter-story drift (local deformation capacity) is especially important when near-fault earthquake ground motions are considered.

This study aims to develop an approximate expression for estimation of maximum inelastic base level drift demand for regular framed multistory buildings subjected to near-fault earthquake ground motions. This approximate expression is intended to provide a quick and rough estimate to be used during the preliminary design of new multistory framed buildings and for rapid evaluation of existing buildings of the same kind using displacement based methods.

Keywords:

STRENGTHENING OF EXISTING REINFORCED CONCRETE FRAMES

ERDEM, Ibrahim
MS
Supervisor: Assistant Prof.Dr. Ugurhan AKYÜZ
Co-Supervisor:

Due to inadequate lateral stiffness many reinforced concrete buildings are highly damaged or collapsed in Turkey. In order to improve the behavior of such buildings and in order to prevent them from collapse, necessary amount of structural walls can be provided. These walls take the most of the lateral loads. The frame of the test specimen (1/3 scaled, 2-story,3-bay) is detailed such that it has the same deficiencies of existing buildings in Turkey. Infill wall is detailed properly. The test specimen is subjected to reversed cyclic quasi-static loading. By means of special transducers, axial force, shear and moment at the base of the columns are measured. Strength, stiffness, energy dissipation and story drifts of the test specimen will be measured.

Keywords:

MECHANICAL PROPERTIES OF CFRP ANCHORAGES

OZDEMIR, Gokhan
MS
Supervisor: Assistant Prof.Dr. Ugurhan AKYUZ
Co-Supervisor:

Due to inadequate lateral stiffness many reinforced concrete buildings are highly damaged or collapsed in Turkey. To improve the behavior of such buildings and to prevent them from collapse, repair and/or strengthening of some reinforced concrete elements is required. One of the strengthening techniques is the usage of CFRP sheets on the existing hollow brick masonry infill. While using the CFRP sheets their mechanical bonds to both structural and non-structural elements are provided by CFRP anchorages. In this study, by means of the prepared test setup, the direct tensile strength capacities of CFRP anchorages are measured. The affects of concrete strength, anchorage depth, anchorage diameter, and number of fibers on the tensile strength capacity of CFRP anchorage are studied. An analytical method is also suggested for the calculation of CFRP anchorage capacities.

Keywords: