AN INVESTIGATION OF ACCURACY OF INERTIAL INTERACTION ANALYSES WITH FREQUENCY-INDEPENDENT IMPEDANCE COEFFICIENTS Yilmazok, &Ouml / zg&uuml / n M.S., Department of Civil Engineering Supervisor: Assoc. Prof. Dr. B. Sadik Bakir November 2007, 79 pages The inertial interaction between the soil and structure alters dynamic response characteristics of a structure due to foundation deformability, such that the flexibility and energy dissipation capability of surrounding soil may lead to a significant increase in period and damping of structural oscillations. The inertial interaction analyses can be accomplished through utilisation of frequency dependent foundation impedance coefficients that are reported in literature for various soil conditions and foundation types. However, such analyses should be performed in frequency domain, and applicable to only cases that linear structural response is considered. Alternatively, equivalent frequencyindependent foundation impedance coefficients can be employed, such that a constant excitation frequency is assumed in calculation of these coefficients. In this study, it is assumed that the fundamental frequency of a fixed-base structure, which can be obtained through employing available empirical relationships or a modal analysis, can be substituted for excitation terms in impedance expressions. The error induced in calculation of peak structural distortions is investigated through comparisons of structural response due to frequency-dependent and frequency-independent foundation impedance coefficients. For analyses, a linear single-degree of freedom oscillator is considered for modeling the structure. The frequency-dependent impedance of a rigid disk foundation resting on elastic halfspace is simulated by a limited number of discrete elements. The response calculations are performed in frequency domain, through employing 72 acceleration records. It is concluded that, the natural frequency of fixed-base building can be considered as effective excitation frequency for calculation of foundation impedance coefficients, when the effect of inertial interaction on structural response is moderate. Through employing equivalent-linear approximation for the structural response, it is shown that the conclusion is also valid in cases that nonlinear structural response is considered. However, when the inertial interaction has more profound effects on the structural response, the use of natural frequency of flexible-base structure, which is calculated iteratively due to its dependence on foundation-impedance factors is recommended.

Concern about effect of explosives effect on engineering structures evolved after the damage of Second World War. Beginning from 90&rsquo / s with the event of bombing Alfred P. Murrah Federal building located in Oklahoma City this concern deepened and with the attack to World Trade Center twin towers on September 11, 2001 it is peaked. Recent design codes mainly focus on earthquake resistant design and strengthening of the structures. These code design methodologies may sometimes satisfy current blast resistant design philosophy, but in general code compliant designs may not provide recognizable resistance to blast effect. Therefore designer should carry out earthquake resistant design with the blast resistant design knowledge in mind in order to be able to select the most suitable framing scheme that provide both earthquake and blast resistance. This is only possible if designer deeply understands and interprets the blast phenomenon. In this study, it is intended to introduce blast phenomenon, basic terminology, past studies, blast loading on structures, blast structure interaction, analysis methodologies for blast effect and analysis for blast induced progressive and disproportionate collapse. Final focus is made on a case study that is carried out to determine whether a regular steel structures already designed according to Turkish Earthquake Code 2007 requirements satisfy blast, thus progressive collapse resistance requirements or not.

Reinforced concrete structures may be subjected to blast loads together with static loads during their service life. Important buildings may be attacked by using explosives as a part of increasing global terrorist activities. Evaluation of blast phenomena for economically and strategically significant buildings is especially important, in order to analyze and design their structural members subjected to air blast loading. Understanding nature of explosions, which are loading characteristics and relation to selected parameters such as explosive type, quantity, and distance, were studied in this thesis. Earlier studies found in the literature survey on explosives, blast, and behavior of structural elements were investigated. Behavior of structures under blast load was described in terms of pressure magnitude, distribution, and reflection phenomena. Simple design, assessment guidelines, and useful charts were developed. A computer program was generated using MATLAB programming language, which automatically generates the air blast pressure versus time data resulting from an air explosion in addition to finite element model formation and dynamic time stepping analysis of a reinforced concrete column. The shear and moment capacities can be calculated and compared against dynamically calculated demand under known axial column force / therefore, vulnerability of a column under blast loading is evaluated. The results of the numerical analyses indicated that failure mechanism of columns is mostly shear failure instead of moment (i.e., plastic hinge and mechanism formation).

Seismic provisions for steel buildings present limiting width-thickness and slenderness ratios for bracing members. Most of these limits were established based on experimental observations. The number of experimental studies is limited due to the costs associated with them. With the rapid increase in computing power / however, it is now possible to conduct finite element simulation of brace components using personal computers. A finite element study has been undertaken to evaluate the aforementioned limits for pin-ended pipe section steel braces. Fifty four tubular pipe brace models possessing different diameter-to-thickness ratios varying from 5 to 30 and slenderness ratios varying from 40 to 200 were analyzed. The effect of cyclic hardening modulus on the response of braces was explored. In all analysis, the models were subjected to reversed cyclic displacements up to ten times the yield displacement. Local buckling was traced during the loading history using a criterion based on local strains. Results are presented in terms of the ductility level attained by the member at the onset of local buckling. It is shown that local buckling of the section is influenced by the diameter-to-thickness and the slenderness ratios of the member. Moreover, the amount of hardening modulus was found to affect the local buckling response significantly. The need to include this material property into seismic provisions is demonstrated. Finally, the hysteretic energy dissipated by the member was quantified for each displacement excursion.

Structures in high seismic risk areas may be susceptible to severe damage in a major earthquake. Structures designed to meet older code requirements may be at even greater risk. When these structures are evaluated with respect to current code criteria, it is observed that they lack of lateral strength and/or ductility. Since safety and economic considerations are major problems, these structures become viable candidates for retrofit and seismic strengthening. For the variety of structures and possible deficiencies that arise, several retrofitting techniques can be considered. Diagonal bracing system is one of the retrofitting techniques and it provides an excellent approach for strengthening and stiffening existing building for lateral forces. Also, another potential advantage of this system is the comparatively small increase in mass associated with the retrofitting scheme since this is a great problem for several retrofitting techniques. In this study, the use of steel bracing for the strengthening of low, intermediate, and relatively high rise reinforced concrete frames are investigated analytically. The ultimate lateral load capacities of the strengthened frames are determined by a load controlled push-over analysis. The post-tensioning effect of preloading is also investigated.

Behavior of lap splices formed by CFRP rolls has been studied. CFRP rolls have been prepared by using CFRP sheets of a certain width. Strengthening methods that use CFRP rolls as reinforcement may require an epoxy anchored lap splice due to the conditions at the strengthening regions. It may not always be possible to strengthen the region by using only one roll fan anchored at both ends, but using two rolls from opposite faces of the member and lap splicing them at the middle so that they act as a single roll. Lap splice behavior can be studied best by using flexural beam bond specimens if the reinforcing material is steel. Therefore, it has initially been suggested that flexural beam specimens reinforced for flexure with CFRP rolls as tension reinforcement can be used in studying the lap splice behavior. However, due to the difficulties encountered in the beam tests, another type of test specimen was introduced, which was a direct pull-out specimen. In this type of test specimen, lap spliced CFRP rolls have been tested under direct tension, in which the tension has been applied by making use of concrete end blocks that transfer the tension to the rolls. Eleven tests have been made in total. Full material capacity of the rolls could not be achieved due to premature failures. However, important conclusions and recommendations have been made for future studies.

In recent years, Artificial Neural Networks (ANN) have become widely popular tools in various disciplines of engineering, including civil engineering. In this thesis, Multi-layer perceptron with back-propagation type of network is utilized in calibration of the finite element model of a long span cantilever through truss called Commodore Barry Bridge (CBB). The essence of calibration lies in the phenomena of comparing and correlating the structural response of an analytical model with experimental results as closely as possible. Since CBB is a very large structure having complex structural mechanisms, formulation of mathematical expressions representing the relation between dynamics of the structure and the structural parameters is very complicated. Furthermore, when the errors in the structural model and noise in the experimental data are taken into account, a calibration study becomes more tedious. At this point, ANNs are useful tools since they have the capability of learning with noisy data and ability to approximate functions. In this study, firstly sensitivity analyses are conducted such that variations in dynamic properties of the bridge are observed with the changes in its structural parameters. In the second part, inverse relation between sensitive structural parameters and modal frequencies of CBB is approximated by training of a neural network. This successfully trained network is then fed up with experimental frequencies to acquire the as-is structural parameters and model updating is achieved accordingly.

About half of the total building stock in Turkey is masonry type building. Masonry buildings in Turkey, especially in rural areas, are constructed without any engineering knowledge mostly by their own residents. They generally have heavy roofs. Masonry type buildings also have thick and heavy wall materials. Heavy roof and wall material generate large inertial forces in the case of an earthquake. Brittle failure of walls leads to total failure of whole system followed by sudden collapse of heavy roof. The aim of this thesis is to understand failure mechanisms of brick masonry walls, prevent their brittle failure and allow the walls to dissipate energy during an earthquake. Furthermore, ultimate capacity increase was also targeted by using low cost and easy to obtain material. In order to find an economical and effective way in strengthening of brick masonry walls in their in-plane direction, steel rebars were used as post-tensioning materials in brick masonry walls and house tests. Springy connections were utilized in the reinforcing and post-tensioning bars in order to prevent early loss of post-tension due to wall cracking or rebar yielding. Separate tests were conducted with and without rebars and springs in order to compare their results. v The test results indicated that the ultimate lateral load capacity of 6m long brick masonry house increased up to about 6 times with respect to its nominal value. Energy dissipation also increased up to about 10 times of the original house. Lateral load capacity increase in 2m long rebar post-tensioned brick masonry walls were measured as about 17 times when compared with the original wall. The energy dissipation capacity was also increased about 30 times the nominal value. A general procedure was developed to assess the vulnerability of single storey masonry houses, which calculates the earthquake demand acting on each wall segment. Comparison of capacity versus demand enables evaluation of wall segments and leads strengthening calculations if necessary. Derived formulas were used to calculate post-tensioning force and design vertical and diagonal rebars. The procedure was demonstrated using properties of an existing house and strengthening cost was found to be about 10% of the building cost. The results of the conducted tests have shown that rebar post-tensioning of brick masonry walls is an effective and cost-efficient way of strengthening the walls in their in-plane direction and can be used as an economical and simple technique for seismically vulnerable masonry houses. Spring based connection detail has improved the post cracking performance of the walls at large deformations by keeping the wall reaction higher after ultimate strength has reached as well as increased the energy dissipation capacity of the walls.

A typical buckling restrained brace (BRB) consists of a core segment and a buckling restraining mechanism. When compared to a conventional brace, BRBs provide nearly equal axial yield force in tension and compression. Buckling restraining mechanism can be grouped into two main categories. Buckling is inhibited either by using a concrete or mortar filled steel tube or by using steel sections only. While a large body of knowledge exists on buckling restrained braces the behavior of steel encased BRBs has not been studied in detail. Another area that needs further investigation is the detailing of the deboding material. For all types of BRBs a debonding material or a gap has to be utilized between the core brace and the restraining mechanism. The main function of the debonding material is to eliminate the transfer of shear force between the core brace and the restraining mechanism by preventing or reducing the friction. A two phase research study has been undertaken to address these research needs. In the first phase an experimental study was carried out to investigate the potential of using steel encased BRBs. In the second phase a numerical study was conducted to study the friction problem in BRBs. The experimental study revealed that steel encased braces provide stable hysteretic behavior and can be an alternative to mortar filled steel tubes. Material and geometric properties of the debonding layer for desired axial load behavior were identified and presented herein.

The purpose of this study is to investigate the influence of material and geometric nonlinearities occurring in beams, columns and walls of RC frame-wall structural systems when undergoing severe ground excitations. For this purpose, a low-rise RC building is considered with and without walls, and the joining beams and columns are designed with the strong-column weak-beam concept. The dimensions, material properties and the reinforcement amounts are calculated in accordance with the values suggested in design codes. Each structure is analyzed for various levels of applied vertical force and change in wall stiffness / where the effect of geometric nonlinearity is considered for each case. Force formulation frame elements with spreading inelasticity over the span are used for the modelling of each beam, column and wall. The coupling of the section forces is obtained by the fibre discretization of the section into several material points. Each section is divided into confined and unconfined regions and appropriate material properties are used for concrete and steel for cyclic loading. Both static pushover and dynamic analyses are performed in order to replicate the worst case scenario for a possible earthquake. From this study, it is concluded that the beams and columns of a frame-wall structural system should be designed carefully for load redistributions resulting from the yielding of the wall in the case of a strong earthquake, thus the design codes should address this situation for both in the retrofit of existing frame buildings with walls and in the construction of new frame-wall type buildings.