Ph.D. Theses

Permanent URI for this collection

https://digitalarchive.library.bogazici.edu.tr/handle/123456789/11708

Browse

Now showing 1 - 20 of 21

Localisation of attenuation equations based on strong ground motion data from dense urban networks
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2023., 2023) Malcıoğlu, Fatma Sevil.; Hancılar, Ufuk.; Şafak, Erdal.
The regional dependency of ground motions underlines the necessity of adopting a more localized approach in the estimation of ground motion parameters (GMPs). This thesis kicks off with a consistency assessment of the five most suitable regional or global ground motion prediction equations (GMPEs) for the interested area and our local database including 6,534 individual horizontal-component ground motions recorded by the ˙Istanbul Earthquake Rapid Response and Early Warning System (IERREWS) network during 78 small-to- moderate earthquakes. The examination uncovers a systematic overestimation tendency in GMPEs. Using the same database, an analysis is conducted to ascertain whether there are any regional azimuth-dependent effects on ground motions. Due to the vagueness of azimuthal effects, the ground motion recording axes are converted to their principal axes, indirectly introducing the azimuthal influence into the analyses. This thesis suggests a station-specific methodology to refine GMP estimations by isolating regional variations. Evaluation of regression outcomes, distance scalings, and residuals guide the selection of the most appropriate functional forms for the empirical equation. Comparisons of the spatial distribution of peak ground acceleration (PGA) for hypothetical and actual earthquakes serve as the basis for the evaluation of the result consistency. Local variations between GMPEs’ estimates and station-specific approximations provide insight into the impact of regional effects on ground motions. The assertion is substantiated by the finding that PGAs calculated from actual earthquake records demonstrate a closer match with station- specific methodology’s predictions, especially in regions with observed local differences.
Real-time structural health monitoring using statistical methods
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2023., 2023) Dar, Emrullah.; Çaktı, Eser.
The detection of structural damage relies on understanding the long-term variation of modal parameters and their relationship to changes in atmospheric conditions. This thesis aims to address this challenge by developing a real-time algorithm for structural health monitoring systems, which are becoming increasingly important. The algorithm uses statistical models developed by analyzing four years of modal frequencies, damping ratios, and mode shapes of Hagia Sophia, a UNESCO World Heritage structure, and their correlation with atmospheric parameters such as temperature, humidity, and wind speed. The algorithm uses four different regression models to predict the modal frequency as a function of the atmospheric conditions and selects the most suitable one. It then compares the predicted and measured frequencies to identify structural anomalies. The algorithm also employs the Modal Assurance Criterion (MAC), Coordinate Modal Assurance Criterion (COMAC), and Enhanced Coordinate Modal Assurance Criterion (ECOMAC) methods to examine the long-term variation of mode shapes. The algorithm is implemented in a user interface software called “AISHM,” which displays the modal parameters and the 3-D animation of the structure in real-time. The software also has the capability to track earthquakes and analyze the structural response in real-time. In summary, this thesis presents a comprehensive approach to real-time structural health monitoring using statistical models and advanced analysis techniques, which can have significant implications for maintaining and preserving historical structures.
Improvement of seismic behaviour of retaining walls by using cushions
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2023., 2023) Tunçay, Gözde Sezgin.; Edinçliler, Ayşe.
Damages and collapses of retaining walls that occur after earthquakes cause loss of life and property and huge economic losses. The use of lightweight material as a cushion behind the retaining wall is one of the improvement methods. The main aim of this thesis is to evaluate the effectiveness and efficiency of a new proposed cushion layer as EPS beads and sand mixture (EPSB) for retaining structures. This cushion material placed behind the retaining wall was used for the first time in the literature. The effects of the mixture ratio and thickness of the EPSB cushion layer under different input motions were investigated. The other cushion layers as EPS geofoam and tire waste and sand mixture (TW) were also used. The performances of three cushion materials were compared with each other under the same input motions. In the experiments, a 1/15 scaled retaining wall model was designed to investigate the effectiveness of the cushion materials on the seismic performance of the retaining wall. Firstly, the experiments for the retaining wall model without cushion material were carried out on the shake table in the laminar box. In the second part, three different types of cushion materials were placed behind the retaining wall by changing thickness, density, and mixture ratios. EPS geofoam was found to provide the highest improvement performance compared to the TW and the EPSB cushions. Evaluation of the test results showed that the cushion type is significantly effective in improving the retaining wall model. Additionally, changes in thickness, density/mixing ratio, and input motions affect the results.
Response analysis and damage mitigation of buried continuous pipes subjected to faulting actions
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2022., 2022) Perdibuka, Dardan.; Edinçliler, Ayşe.; Uçkan, Eren.
Buried continuous steel pipelines are critical lifelines failure of which under fault rupture incidents may lead to significant and deteriorating environmental and socio economic outcomes. Proper understanding and estimation of the mechanical behavior of buried steel pipes under such geohazards and investigation of means of mitigating these deleterious effects is of paramount importance. This thesis aimed at developing rigorous and simplified numerical models of the problem to realistically simulate the behavior of buried continuous pipes under strike-slip fault rupture-induced permeant ground deformations. The response of buried pipe cases under the fault load was investigated with respect to the variation of fault crossing angle (β) and pipe wall thickness (t). The second phase of this dissertation involved the investigation of the effect of four mitigation techniques to protect the buried pipe against fault rupture-induced damages. Lastly, a case study involving the evaluation of the effect of using CFRP wraps on the response of Thames Water Pipe which suffered great damage during the devastating 1999 Izmit is presented. The outcomes of this thesis indicate that the performance of the pipeline is sensitive to the variation of fault crossing angle and pipe wall thickness, increasing both parameters lead to overall improved pipe performance. Results indicate that all mitigation approaches offer certain degrees of improvement, where most effective mitigation approach is the wrapping of the pipeline surface with CFRP wraps while the use of controlled-low strength material was the least effective approach. Comparison of simplified and rigorous numerical models revealed that a good agreement exist between the approaches. Lastly, evaluation of the response of Thames Water Pipe protected using CFRP indicates that despite the considerable reduction in stresses and strains complete avoidance of failure for this particular case does not seem to be attainable.
Constraining source properties of the 1509, 1766 and 1894 Istanbul earthquakes
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2022., 2022) Yenihayat, Nesrin.; Çaktı, Eser.; Şeşetyan, Karin.
In this thesis, stochastic simulations were performed for the 10 September 1509, 22 May 1766, and 10 July 1894 Istanbul earthquakes to constrain their source parame ters with a detailed characterisation and classification of associated damages and their distributions. Damage information was obtained as spatial distributions of observed data assembled from various sources and interpreted on the MMI scale. Using a finite fault simulation approach, numerous scenarios were modeled with calibrated and validated source, path and site parameters. In calibration, simulations were performed for 59 recordings of the 26 September 2019 Silivri (Mw=5.8) earth quake. In order to test our model on a larger Marmara earthquake recordings obtained at 19 stations of the 17 August 1999 Kocaeli (Mw=7.4) earthquake were also simulated. Through an iterative selection of source parameters and comparison of simulation based synthetic intensity distributions with intensities assessed from actual damage, evaluations of source parameters of three earthquakes were made. Findings indicated that most likely the magnitude of the 1894 Istanbul earthquake was Mw=[6.8 to 6.9], and its epicenter was either at the midpoint of the Prince Islands segment with a 42 km rupture or at the intersection of the Prince Islands segment with the Izmit segment as a 60 km total rupture. The 1766 earthquake likely had a rupture length of about 80 km along the Central Marmara Fault with a magnitude of Mw=[7.0 to 7.1]. Our findings suggested that the 1509 earthquake most probably occurred as a combined rupture of the Prince Islands segment and the Central Marmara Fault, and created a 120 km rupture with a magnitude of Mw=7.6.
Spatial variation of strong ground motion
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2009., 2009.) Harmandar, Ebru.; Durukal, Eser.; Erdik, Mustafa.
Earthquake ground accelerograms measured at different locations along a large engineered structure could be significantly different. This has led to considerable research in the last decade on the modeling of the spatially varying earthquake ground motion. The spatial variability of strong ground motion incorporates the effects of wave propagation, amplitude variability and phase variability, as well as the local site effects on the motion. This variation of ground motion could have the possibility to cause important effect on the response of linear lifelines such as long bridges, pipelines, communication systems, and should preferably be accounted for in their design. The objective is to evaluate and improve existing spatial variation quantification relationships by studying data available from different networks; investigate the possibility of employing functional forms for the characterization of spatial variation of ground motion in the assessment of strong ground motion distribution. This thesis focuses on studying on the spatial variability of ground motion using strong ground motion measurements. A rational and rigorous methodology for the interpolation of measured ground motion from discrete array stations to be used in the bias adjustment of the theoretical shake map assessments with the empirical ground motion measurements is developed. The generation of the estimated maps of shaking after an earthquake is often influenced by the limited number of sensors and/or difficulty of monitoring at inaccessible locations that impacts the collection of desired information. This gap in information can be filled through the estimation of missing information conditional upon the measured records. Methodology is presented for estimating properly-correlated earthquake ground motion parameters; herein peak ground acceleration (PGA), at an arbitrary set of closely-spaced points, compatible with known or prescribed ground motion parameters (PGA) at other locations. The variation of ground strain due to wave propagation, site response and loss of coherence is investigated. This study concentrates on the stochastic description of the spatial variation, and focuses on spatial coherency. The estimation of coherency from recorded data and its interpretation are presented. Coherency model for Istanbul for the assessment of simulation of spatially variable ground motion needed for the design of extended structures is derived. In addition to the realistic characterization of spatial variation, simulation of spatially variable earthquake ground motion is another essential part of the examination of the effects of spatial variation, especially for extended lifeline structures. This thesis concludes with the generation of earthquake ground motion compatible with prescribed target-response spectrum and their coherencies are consistent with a given spatial coherency function for a finite array of ground surface stations.
Nonlinear fluid-structure interaction for multi-dimensional seismic analyses of liquid storage tanks
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2009., 2009.) Özdemir, Zühal.; Şafak, Erdal.; Souli, Mhamed.
Real nonlinear behavior of liquid storage tanks includes many complexities which are caused by material yielding, large amplitude free-surface sloshing, non-linear fluid structure interaction, high deformations of tank base and shell, out-off round distortions of the tank shell, soil-tank interaction, successive separation and contact between tank base and foundation and plastic rotations of tank base plate. These nonlinear behavior mechanisms result in different failure modes such as buckling at the tank shell (elephant foot buckling or diamond shape buckling), separation of the junction between the base plate and tank wall due to high joint stresses, uneven settlements at the tank base and rupture of the anchors. The algorithm to be employed for the seismic analyses of tanks should account for these nonlinearity effects for the accurate description of the performance of tanks during earthquakes. In this thesis, fluid-structure interaction algorithm of finite element method which can take into account the effects of geometric and material nonlinearities of the tank and nonlinear sloshing behavior of contained liquid is utilized to evaluate the actual behavior of steel cylindrical ground supported liquid storage tanks when subjected to realistic base motions. Since seismic design codes generally define ground shaking in the form of an acceleration response spectrum, earthquake ground motions is selected and processed using spectrum matching techniques in time domain to be compatible with the Turkish Seismic Design Code (2007) spectra. In addition to two horizontal components of ground motion, the vertical component is also taken into account in order to determine relative importance of vertical ground motion on the behaviors of anchored and unanchored tanks. In order to clarify the key question of tank problems whether anchoring would prevent earthquake damage to the tank, numerical analyses are carried out on the same tank model having two different support conditions: anchored and unanchored. The consistency of the provisions presented in current tank seismic design codes and finite element method analysis results are evaluated and recommendations on seismic tank design codes are presented.
Tools and techniques for real time modal identification
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2009., 2009.) Kaya, Yavuz.; Şafak, Erdal.
Framework of the structural health monitoring (SHM) is to detect and locate the damage in the structure. Real-time identification of modal properties of a structure is very important in SHM because the changes in modal properties during its service life may be related to the damage on the structure. Therefore, it is obvious that modal properties should be monitored and identified accurately in real-time in order to have a good estimate in SHM. A new real-time software together with new real-time tools/techniques have been developed in this dissertation to be used in SHM to identify the modal properties of the structure in real-time. In the new proposed tool the real-time modal frequencies are estimated by using basic signal processing tools such as baseline correction, band-pass filtering, windowing, FFT, and smoothing while the real-time damping is estimated with half-power bandwidth technique and/or Logarithmic decrement technique. The developed real-time software, KOERIMIDS, has been tested with the ambient vibration data set recorded from Hagia Sophia Museum. Modal properties of the structure have been identified successfully in real-time. Results of the Hagia Sophia test have been compared with the previous studies conducted by different researchers. Comparison shows that the results of the KOERI-MIDS are in good agreement with that of the previous studies.
Dynamic shear amplification in seismic response of structural wall systems
(Thesis (Ph.D.)-Bogazici University. Kandilli Observatory and Earthquake Research Institute, 2008., 2008.) Celep, Umut Utku.; Aydınoğlu, M. Nuray.
Previous research indicates that, shear force demand in yielding walls are not proportional to the design moments calculated by code procedures and higher shear force demands develop along the wall with respect to code predictions as a consequence of the higher mode effects after the plastic hinge formation at the base of the wall. 2007 version of the Turkish Seismic Design Code takes the dynamic shear amplification phenomena into account with a constant base shear amplification factor of 1.5 regardless of the first mode period and ductility level of the wall. However, results obtained from extensive non-linear time history analyses performed on generic walls in this study indicate that dynamic shear amplifications increase with increasing first mode period, Strength Reduction Factor (R) and ground motion intensity. A dynamic base shear amplification relationship as a function of the first mode period and strength reduction factor has been proposed for the Turkish Seismic Design Code (2007), based on the regression analysis of the non-linear time history analysis. A story shear force profile has been suggested for the Turkish Seismic Design Code (2007), which is intended for not only preventing shear failures at the base but also along the height of the wall. As a side product of the nonlinear time history analyses, a moment profile has also been proposed for use in the Turkish Seismic Design Code (2007). A modal decomposition technique is presented in this study for demonstrating the effects of the higher modes on the dynamic shear amplification phenomenon.
Deformation based seismic design of pile supported marine facilities
(Thesis (Ph.D.)-Bogazici University. Kandilli Observatory and Earthquake Research Institute, 2008., 2008.) Polat, Şamil Şeref.; Aydınoğlu, M. Nuray.
Pile supported marine structures with batter piles comprise a considerable share in the modern marine structure stock built in seismic zones. Their stiff nature offers significant advantages to structural engineers in resisting non-seismic loads such as, berthing, mooring etc. On the other hand the poor performance of marine structures supported by batter piles in recent earthquakes has revealed certain disadvantages of these systems in resisting seismic loads. The general design approach for pile supported marine structures is to ensure that the cap-beam and the deck system will remain elastic and the yielding will occur either at the pile-to-cap beam connection or along the pile itself. Traditionally those structures were designed with force-based design methods to withstand seismic forces reduced by response modification factors or to a force equal to a fraction of the total weight of the structure. The past research in last decade have shown that the poor behavior of batter piles is mainly related to this force-based approach, which lack to identify the problems associated to post-yield behavior of these piles. When batter piles yield in tension, either in the form of pile-to-cap-beam connection or pile pull-out of soil pile-cap starts to pole vault over the compression piles as the structure deforms laterally. As the structure rises, substantial tension forces are developed both in the vertical and orthogonal batter piles and create additional shear and moment to the cap-beam. The non-linear analysis performed on generic pier frames in this study revealed that substantial amplifications in section forces have been observed at the pile-cap with decrease in strength and increase in batter. Even though section compactness is a well known requirement in steel design most of the modern marine structure design codes does not provide a compactness criteria. When the connection of the tension batter pile is designed to develop high axial forces, the compression piles with non-compact steel section have a tendency to yield under the action of earthquake induced bending moments and high compression forces at the pile-soil interface. The inelastic response history analysis performed on generic pier frames indicate formation of inelastic local buckling on compression piles results in partial or total collapse of the structure. Design recommendations are provided based on the results of the performed nonlinear analysis.
Deformation styles and rates along the North Anatolian Fault zone in the sea of Marmara based onshore-offshore seismic, geodetic and geologic data
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2021., 2021.) Coşkun, Zeynep.; Pınar, Ali.
The aim of the thesis is to make a contribution on the long lasting debate on complex fault geometry along the northern branch of the NAF beneath the Sea of Marmara, using seismological, geodetic and geologic data. The focal mechanisms of: (1) small to moderate size events are obtained by CMT inversion technique of Kuge (2003), using onshore waveform data from 2002-2015, (2) micro-earthquakes are obtained using technique of Horiuchi (2015), using offshore waveform data recorded by 15 OBS stations from 2015-2016. The geodetic horizontal crustal strain rates are determined at equally spaced grid points by interpolating northing and easting components of the 112 GPS vectors from 1994-2013. The strain ε̇ and moment rates Ṁ are calculated by Kostrov’s (1974) summation method later modified by Ward (1994). The results indicate that extensional and strike-slip style dominates the region, while compression features are rare. Significant elongation is observed in Çınarcık Basin (CB) and the area between Marmara Island and Central Basin (CeB). Compressional or transpressional features is derived to the west of Marmara Island and Ganos and in the Central segment extending from CeB toward CB. The sparse seismicity beneath the Kumburgaz Basin yields the lowest ε̇ estimated in the region, 11x10-8 /y, supporting the idea that this section could be locked and accumulating strain. The Ṁ estimation results for each segment show that Ṁseis, calculated for instrumental period, are greatly lower than Ṁgeod and Ṁgeol. This feature can be interpreted in two ways: (1) action of aseismic strain release (creeping), (2) strain accumulation along fault segments is underway and only small portion of the accumulated seismic energy is released by small magnitude events. ε̇geod results point out the highest values, 24x10-8 /y, in CB, while the lowest values, 11x10-8 /y, are observed in Central Marmara. The highest ε̇ in both edges of the fault segment in CB can be indicative of steadily creeping fault segment. Vice versa lower strain rates in Central Marmara region suggest that this segment of NAF is locked.
A seismic vulnerability analysis procedure for urban loss assessment
(Thesis (Ph.D.)-Bogazici University. Kandilli Observatory and Earthquake Research Institute, 2008., 2008.) Tüzün, Cüneyt.; Aydinoğlu, M. Nuray.
The assessment of seismic vulnerability of buildings in seismically active urban areas is of great importance in terms of engineering, economical and social aspects. In this study an analytical procedure has been developed in order to obtain the vulnerability functions of existing buildings. The building information included over all geometry of the structure, as-built dimensions and configuration of the structural members, complete with reinforcement details of beams and columns, which are extracted from the design dossiers available in the Bolu municipality archives. Totally 120 buildings, 20 from each class, are classified depending on their number of stories ranging from 2 to 7 story and nonlinear response history analysis has been performed in order to obtain damage distributions by using 20 spectrum compatible ground motions. In the first chapter a brief explanation of the role of loss estimation studies in urban planning, urban disaster management and mitigation has been given. In the second chapter, existing building vulnerability relationships have been evaluated and particular examples of studies that have made important contribution to the development are briefly cited. In the third chapter, basic steps of a standard derivation of analytical vulnerability functions are identified and the importance of each step in the process has been explained. In the fourth chapter, the procedure based on real building data has been explained in detail including the ground motion characterization, properties of building data, analytical methods used in the analysis, damage definition and quantification, statistical process and the determination of the vulnerability parameters. Results are presented in the form of lognormal vulnerability curves with respect to spectral acceleration or spectral displacement. In the fifth chapter, the conclusions are presented and the contribution of the study to the existing knowledge in the field has been evaluated
Evaluation of practice-oriented nonlinear analysis methods for seismic performance assessment
(Thesis (Ph.D.)-Bogazici University. Kandilli Observatory and Earthquake Research Institute, 2008., 2008.) Önem, Göktürk.; Aydınoğlu, M. Nuray.
In the last decade, nonlinear static analyses based on pushover analysis have been developed as a simplified nonlinear analysis tool within the context of performance based design approach (ATC 40 and FEMA 356). Since nonlinear static analyses provide designers a practical analysis approach in estimating inelastic seismic demands, these methods have been widely used in engineering practice. On the other hand, recent research have clearly shown that simplified nonlinear static analyses, which consider single mode behavior of the structures, have serious limitations for high-rise buildings or buildings irregular in plan, where higher modes effects become important. In order to overcome these limitations and to enhance the feasibility of the pushover analysis in practice, a number of multi-mode pushover analysis methods have been developed. It should be noted that pushover analysis has not been provided with a firm theoretical basis and those methods are therefore based on various assumptions. In this study, development and codification of nonlinear static analysis as a tool for performance based assessment have been summarized. Piecewise linear representation of single-mode pushover analysis, which provides a non-iterative pushover analysis technique with an adaptive load or displacement pattern, has been presented in detail. A number of multi-mode pushover analysis methods have been investigated in detail and classified with respect to their assumptions. The emphasis of this study is to evaluate the validity of those assumptions and their limitations in terms of practical applicability. In addition, a parametric study is carried out in order to evaluate and understand the limitations of single-mode and multi-mode pushover analysis methods based on various assumptions. It has been observed that some multi-mode pushover analysis methods deal with estimating only structural capacity, resulting in a conventional pushover curve where higher modes effects are somehow considered. Thus these multi-mode pushover analysis methods can be regarded only as capacity estimation tools. However, the main objective of the nonlinear static analysis should be the estimation of the seismic demands under a given earthquake ground motion. It is interesting to observe that the number of multi-mode pushover analysis methods achieving this objective is very limited. Determination of relative modal contributions at each pushover step with an appropriate modal scaling procedure is a critical point in a multi-mode pushover analysis As a result of the investigation of multi-mode pushover analysis methods, it has been observed that there are mainly two types of modal scaling procedures generally adopted: (a) scaling based on instantaneous inelastic spectral displacements, (b) scaling based on instantaneous elastic spectral displacements or pseudo-accelerations. It has been identified that multi-mode pushover methods adopting modal scaling procedure based on instantaneous elastic spectral quantities would not work when P-delta effects are considered. The effectiveness of multi-mode pushover analyses has been tested for reinforced concrete frame and dual systems by comparing the results obtained from inelastic time history analysis (ITHA). Analysis results indicated that multi-mode pushover analyses, which combine multi-mode effects at each pushover step, provides relatively good estimates of inter-story drift and plastic rotation demands in the lower and middle stories of taller frames. At the upper story levels, where higher mode effects are significant, Incremental Response Spectrum Analysis (IRSA) developed by Aydınoglu (2003) and Modal Pushover Analysis (MPA) developed by Chopra and Goel (2001) give more accurate results as compared to the other methods. It has been observed that when P-delta effects are included in the analyses, the discrepancy between the results obtained from ITHA and all pushover analyses tends to increase as compared to the case without P-delta effects. For dual systems, multi-mode pushover analyses, which combine multi-mode effects at each pushover step, predicts reasonably well the changing height-wise variation of plastic rotation demands in the beams with building height, particularly for dual systems with smaller wall shear ratio. IRSA significantly predicts much more accurate plastic rotation estimates with respect to all other multi-mode pushover methods. Single-run multi-mode pushover analysis methods with single-load or single-displacement patterns based on combined multi-mode loading significantly underestimate shear force demands in the shear wall elements. Additionally, it has been observed that multi-mode pushover analysis methods provide much more accurate estimate of plastic hinge rotations and their locations at the base of the shear walls as compared to FEMA 356 lateral load distributions. Single-mode adaptive pushover analysis can predict plastic rotation demands accurately at the base of the shear walls in spite of the fact that only single mode was considered, whereas invariant single-mode pushover analysis cannot predict. This shows that adaptive pushover analysis provides a more reliable analysis technique, which is able to capture changing dynamic characteristics of dual systems and eventually plastic rotation demands at the base of the shear walls.
Factors affecting site response analysis
(Thesis (Ph.D.)-Bogazici University. Kandilli Observatory and Earthquake Research Institute, 2009., 2009.) Tönük, Gökçe.; Ansal, Atilla.
The objective of a site response analysis is to estimate free-field ground shaking characteristics during an earthquake for a specific hazard level and set of site conditions. The mandatory components for a site response analysis are: one or more design earthquake records with representative acceleration time histories, an idealization of the soil-rock system at the site of interest, and a scheme to generate response solutions to simplified assumed wave fields in other words appropriate modeling of the soil behavior. Normally, the free-field ground response is presented in terms of either response spectra or the variation of acceleration or velocity with time. The study aims to review and improve different components of site response analyses in order to achieve a robust methodology for more comprehensive and realistic assessment. The effects of input acceleration time histories, the applied numerical methodology, stress and frequency dependence and nonlinear site response analysis were reviewed and methodologies were suggested based on case studies. Site response of layered soil deposits was analyzed using equivalent linear and modified equivalent linear schemes. The developed methodology would be utilized to estimate earthquake characteristics on the ground for site specific investigations based on probabilistic earthquake hazard assessment. Within this perspective, site response analysis was studied with respect to (a) the determination of different scaling parameters including derivation of attenuation relationships for these parameters, (b) the evaluation of scaling parameters with respect to magnitude and distance ranges, (c) the methodology of selection and scaling of input acceleration time histories for site response analyses, (d) the methodology for selection of ground motion parameters from site response analysis as design or damage parameters for various earthquake engineering analysis such as liquefaction susceptibility, microzonation, vulnerability assessments for buildings and pipeline networks, (e) the methodology for confining stress and frequency dependence of modulus reduction and damping in equivalent linear site response analysis, (f) the review concerning the available equivalent linear site response analysis models and software, (g) formulation of modified version of Shake91 to account for stress and frequency dependency, (h) comparison of results with modified Shake91 based on selected borings, and (i) the review concerning nonlinear models for site response analysis.
A parametric study for the characterization of site amplification
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2020., 2020.) Fercan, Nazife Özge.; Şafak, Erdal.; Ansal, Atilla.
In earthquake engineering, the approximation of site amplification by using practical ways has been an important issue. Various site parameters were proposed and applied in the engineering practice. Among these, time averaged shear wave velocity for the top 30 m, Vs30, and fundamental frequency, f0, have been used widely. In this study, we investigated the reliability of Vs30 parameter, and the performance of alternative time averaged shear wave velocities (e.g., Vs40, Vs50, etc.) and shear wave travel times (Ttz) at various depths for the estimation of site amplification. For the same bedrock depth, we considered 17 shear wave velocity profiles, changing from convex (i.e., the velocities changing faster near the surface and slower near the bedrock) to concave (i.e., the velocities changing slower near the surface and faster near the bedrock). We divided the soil media, first into layers with equal thickness, and then into layers with equal wave travel times. For each layering type and soil profile, we calculated the site amplification factors and fundamental frequencies, and studied their correlations with time averaged shear wave velocities (Vsz) and wave travel times (Ttz) for different depths, z. We have also investigated the correlation of site amplification factors, surface PGAs (Peak Ground Accelerations), and fundamental soil frequencies (f0) for each case. We have identified the optimal averaging depths for the averaged shear wave velocity and the wave travel time to characterize site amplification. The study showed that there is a sharp change in the correlations when switching from convex to concave profiles. By gradually increasing the bedrock acceleration levels, we have also studied the nonlinear soil response and its correlations with linear soil response. We presented guidelines to estimate nonlinear soil amplification factors and fundamental frequency from the linear ones. Considering that the linear fundamental frequency and amplification can easily be calculated from field tests (e.g., ambient noise measurements for f0 detection), these guidelines provide a useful tool to estimate nonlinear ones.
A proposed ground motion selection and scaling procedure for nonlinear response history analysis
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2016., 2016.) Zengin, Esra.; Akkar, Sinan.
With the advancement in performance-based earthquake engineering, nonlinear response history analysis of structures has become more common in recent years. The selection and scaling of ground motions for use in nonlinear response history analysis is one of the most critical steps in performance-based seismic assessment procedures. This study presents ground motion selection and scaling procedure that addresses the uncertainty in the spectral demand with the preserved dispersion within the ground motion set. The candidate ground motion sets are constructed based on dispersion statistics about the target spectral demand. The optimum ground-motion set is linearly scaled by using an optimization algorithm that minimizes the error between scaled median and target spectra. The scaling stage ensures that the median record spectrum provides a reasonable match to target median in a previously defined period interval. This procedure allows performing further modification on each scaled ground motion in order to match the target variance of the scenario-based spectrum. In this study, a novel probabilistic framework is presented to propagate the uncertainties in both ground motion intensity and the structural response on fragility curve estimations. To investigate the effects of uncertainties on seismic damage estimations, the results of this study are compared with those obtained by the conventional fragility curve approach.
Use of distinct element method in the assessment of earthquake behavior of masonry structures
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2014., 2014.) Saygılı, Özden.; Çaktı, Eser.
There are a number of masonry structures in Istanbul and in other cities of Turkey that suffered severe damage from earthquakes. As they will continue to get affected by the earthquakes, more research is needed to assess their seismic dynamic behavior particularly in the states of large deformations/damage and collapse. This still remains as a challenge in spite of significant developments in understanding factors affecting the seismic resistance of masonry structures. The present thesis deals with non-linear dynamic analysis of masonry structures modelled through distinct element methodology. First, a masonry mosque has been built at 1:10 reduced scale and tested by subjecting it to a sequence of earthquake excitations on the shake table in three phases: test of the base-isolated model, of the model as it is and that of the strengthened model. The results of these three phases were used in the calibration/validation of the numerical model developed by distinct element approach. It has been concluded that the methodology and the elements developed in this stage are good enough to be employed in the investigation of real masonry structures. In the second stage, three masonry minarets in Istanbul were studied under sine waves (velocity amplitude range: 10 cm/s – 100 cm/s; frequency range: 0.1 Hz – 13 Hz) and under real and simulated earthquake ground motion. The deformation levels and patterns induced in the minarets and the energy balance in the system are investigated to analyze the damage processes.
New techniques in dynamıc analysis of structures: Spectral element method, travelling wave method and enenrgy flux
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2018., 2018.) Çağlar, Nilgün Merve.; Şafak, Erdal.
The Spectral Element Method (SEM), Travelling Wave Method (TWM), and the Energy Flux are robust techniques to calculate the dynamic response of the structures, which are all based on the exact solution of the governing diﬀerential equations. In the formulation of these methods, either elementary or higher order element theories can be adopted. Since they use the exact solution, the inertia terms are implemented prop erly. An element without any discontinuity can be modeled as a single element. Thus, the number of degrees of freedoms (DOFs) decreases considerably, providing signiﬁcant reduction in the problem size and the computation time. High-frequency wave modes, which are the modes more sensitive to small changes in the dynamic characteristics of the structure, can be captured more accurately with these methods. The matrix equa tions for the dynamic response of two and three-dimensional structures can easily be formed by assembling the element response matrices derived from the SEM, TWM and Energy Flux approach. All the formulations are given in frequency domain. There fore, damping and SSI (Soil-Structure Interaction) eﬀects can be incorporated more accurately, because many element-level damping properties and foundation impedance functions are frequency dependent. The propagation path of the disturbance within the structure, as well as the dissipated and reﬂected waves and energies can be tracked. This gives more insight into the dynamic behavior and the energy absorption capacity of the structure. Since they are very accurate in high frequencies, these methods also provide more powerful tools for system identiﬁcation and damage detection. Most of the small and invisible damages in structures are hidden in high frequencies.
Correlations between ground motion intensity measures and structural response parameters through nonlinear dynamic analyses
(Thesis (Ph.D.)-Bogazici University.Kandilli Observatory and Earthquake Research Institute, 2009., 2009.) Hancılar, Ufuk.; Durukal, Eser.
Earthquake hazard and risk assessment for Turkey
(Thesis (Ph.D.)-Bogazici University. Kandilli Observatory and Earthquake Research Institute, 2010., 2010.) Demircioğlu, Mine Betül.; Erdik, Mustafa.
Using a GIS-environment to present the results, seismic risk analysis is considered as a helpful tool to support the decision making for planning and prioritizing seismic retrofit intervention programs at large scale. The main ingredients of seismic risk analysis consist of seismic hazard, regional inventory of buildings and vulnerability analysis. There are two main objectives of this thesis. The first objective is the assessment of the national earthquake hazard based on the next generation attenuation (NGA) ground motion prediction models and comparisons of the results with the previous models. The second objective is an evaluation of seismic risk based on a probabilistic intensity ground motion prediction for Turkey. According to the macroseismic approach of Giovinazzi and Lagomarsino (2005), two alternative vulnerability models have been used to estimate building damage. The vulnerability and ductility indices for Turkey have been taken from the study of Giovinazzi (2005). These two vulnerability models have been compared with the observed earthquake damage database. A good agreement between curves has been clearly observed. In additional to the building damage, casualty estimations based on three different methods for each return period and for each vulnerability model have been presented to evaluate the earthquake loss. Using three different models of building replacement costs, an average annual loss (AAL) and probable maximum loss ratio (PMLR) due to regional earthquake hazard have been provided to form a basis for the improvement of the parametric insurance model and determination of premium rates for compulsory earthquake insurance in Turkey.

Browse

Recent Submissions