پژوهشنامه زلزله شناسی و مهندسی زلزله
سال شانزدهم شماره 3 (پیاپی 62، 1392)

In the event of an earthquake that is an unusual and reciprocal loading, it’s permitted in many structures, according to their importance and ductility, to go to their inelastic range of behavior and dissipate some of the input energy in this zone through nonlinear hysteresis loops. Ductility demand should be determined on the members that yielded, so nonlinear analysis is needed. For nonlinear dynamic analysis of structures and determining the ductility demand on the members, and for analyzing the structures in which seismic isolator or damping elements are used, it is needed to provide sufficient number of suitable accelerogram, moreover. With increasing of the knowledge and the interest for using performance design and assessment of structures, using time history analysis is on the rise. In many areas, there is no suitable recorded accelerogram, but there are historical evidences for large earthquakes. Therefore, we need to use scaled or simulated accelerogram for engineering purposes. In the building regulations, it is allowed that simulation of ground motion can be used as a tool for producing accelerogram. A favorite method in earthquake engineering to produce accelerograms is the finite-fault stochastic method. In finite-fault modeling of earthquake ground motion, a large fault is divided into N sub-faults, where each sub-fault is considered as a small point source. In the point-source method for simulating a record, first, a white noise is generated for duration given by duration of the motion. This noise is then windowed. The windowed noise is transformed into the frequency domain. The spectrum of noise is normalized by the square-root of the mean square amplitude spectrum. The normalized spectrum is multiplied by the ground motion spectrum. The resulting spectrum is transformed back to the time domain. Two types of window function that generally use in the stochastic methods are Saragoni-Hart and Boxcar window. Therefore, it may affect [or may be affected by] the response of structures. One of the essential characteristics of the method is that it distills what is known about the various factors affecting ground motions (source, path, and site) into simple functional forms. This provides a means by which the results of the rigorous studies reported in other papers can be incorporated into practical predictions of ground motion. Since the dynamic time history analysis results, depending on the type and content of input information, it's required to consider good accuracy in selection of data. In this study, it has investigated the effect of window functions used to shape white noise on the ductility demand of single degree of freedom systems, showing that the types of window functions do not have any effect on the response of single degree of freedom systems. Besides, it has investigated the effects of time step, used to produce white noise, on the ductility demand and pseudo acceleration response of single degree of freedom systems, showing that the time step equal to 0.02 produces higher ductility demand and lower pseudo acceleration response in the frequencies higher than 2 Hz compared with 0.005 and 0.01.

In seismic hazard assessment for estimating damage function as well as linear and nonlinear response spectrum, we need to estimate seismic strong ground motion. For this purpose, we need ground motion prediction, which can be derived from stochastic or empirical methods. These methods lead to relationships called attenuation relationships. These attenuation relationships predict ground motions by means of seismic source characteristics such as magnitude and fault mechanism, and local parameters such as distance and soil type. Today, many attenuation relationships are presented by researchers. However, their results have conspicuous discrepancy even for the same region. Considering the increasing number and complexity of ground-motion prediction equations available for seismic hazard assessment, there is a definite need for an efficient, quantitative, and robust method to select and rank these models for a particular region of interest. Considering frequent seismic and destructive earthquakes in Iran, there is Significance appropriate Attenuation relation for a particular target area. Selecting the best model for target regions can significantly reduce epistemic uncertainty. This paper uses statistical methods to study goodness of fit 5 candidate attenuation relationships with observed data earthquake 11 August 2012 in Varzaghan. Five attenuation relationships studied [1-5] including Iran's relations with the world. To analyze the statistical distribution of the data set, statical graphics and goodness-of-fit of measurement are used. We show how observed ground-motion records can help to guide this process in a systematic and comprehensible way. A key element in this context is a new, likelihood based, goodnessof- fit measure that has the property not only to quantify the model fit, but also to measure in some degree how well the underlying statistical model assumptions are met. By design, this measure naturally scales between 0 and 1, with a value of 0.5 for a situation in which the model perfectly matches the sample distribution both in terms of mean and standard deviation. The ground motion models are four different classes of quality. Besides, by using statistical tests, the weight of each relation is obtained for using in seismic hazard analysis.

Reinforcement of soil structures such as slopes and walls has become an accepted engineering practice in the last three decades. This paper focuses on reinforced slopes; in particular, the required strength of reinforcement and its length necessary to avoid collapse during seismic events in special factor of safety (FS). The seismic influence is substituted with a quasi-static horizontal force. The reinforcement considered here is of a geosynthetics type (geogrids or geotextile), and it contributes to the stability of the structure only through its tensile strength (the reinforcement's resistance to shear, torsion and bending is negligible). Calculations were performed assuming uniform distribution of reinforcement strength through the slope height, and assuming the Mohr-Coulomb failure criterion holds for the soil. The technique of calculations is based on the kinematic theorem of limit analysis. This theorem states that the rate of work done by traction and body forces is less than or equal to the rate of energy dissipation in any kinematically admissible failure mechanism. Algorithms were written, based on two main mechanisms; (a) several horizontal hexagonal blocks and one pentagonal block at the bottom for overall failure mechanism, and (b) Two pentagonal blocks for direct sliding mechanism. Numerical optimization methods were used to determine FS and critical failure surface. Required reinforcement strength calculated with this assumption that the reinforcement fails by plastic flow referred here as tensile rupture. This happens only when the reinforcement is of sufficient length. The length of reinforcement was also calculated, based on collapse mechanisms that include rupture in some layers and pull-out in others, and the last set of calculations for required length was performed assuming a different mode of failure: direct sliding. For each mechanism, a non-linear equation and a numerical method of optimization was used for solving the equations. The results obtained in this study are presented as dimension less charts which can be used in design, and to evaluate the effect of different parameters such as geometrical parameters of the slope, soil strength, and seismic coefficient on the reinforced soil slopes stability. Obtained results may be applicable to walls, although design of walls requires consideration of additional failure modes not presented here. The results clearly indicate how the required strength increases with decrease in the internal friction angle, increase in the slope inclination angle and increase in the seismic coefficient. The required length is almost in dependent from slope inclination angle in direct sliding mechanism.

Since buildings are individually built close to each other, seismic excitation is caused the buildings experience lateral displacement and collides with each other due to different material, height and vibration modes. Structural pounding, which may occur because of earthquake between two adjacent buildings with different dynamic characteristics, has been an interesting research topic during the last few decades in field of earthquake engineering. Many researchers have intensively studied building pounding, impact and dissipated energy between adjacent buildings during earthquakes. In order to calculate the impact and energy absorption, numerical investigations need to have an unreal link element, which is involved spring and dashpot and is located at the conjunction top level of buildings. Recently, several mathematic equations have been justified to determine impact value and energy dissipation during collision. Most of researchers have used numerical methods to simulate the problem of earthquakeinduces pounding between adjacent buildings. Investigation of building pounding can be addressed in two different paths: experimental analyses and analytical analyses. To measure the impact force during collisions and lateral displacement of adjacent structures, software used need to define a specific link element at the connection level between the buildings analyzed. These link elements can be significantly different so as to insure a complete agreement between analytical and experimental results based on type of link elements. The mathematical equations correspond to the modeling by distinct link elements could be calculated by different approaches. The main different concepts used on link elements correspond to appropriate use of gap, spring and damper in the link elements. As periods of the adjacent colliding buildings are conceptually different, the link elements should be able to allow and translate the different behavior of buildings during seismic excitations. In this study, two new mathematical equations are proposed to measure the impact force and energy dissipation. The results based on the proposed equations are compared with those of the available equations. Since there is a need to have a reference curve to select impact velocity based on coefficient of restitution, several impact velocities and CRs were evaluated. By using the latter curve, results could be evaluated. Finally, based on coefficient of restitution and using a steady-state response of single degree of freedom system due to the external force, a new equation of motion is suggested to estimate the impact damping ratio.

In this paper, seismic design criteria of pipelines, especially for buried gas pipelines from different codes, guidelines and researches are reviewed and compared. For buried or above ground pipelines, both seismic wave propagation, sometimes called as transient ground deformation or TGD, and permanent ground deformation or PGD are important, but for above ground pipelines, like other above ground structures, seismic wave propagation is more critical than PGD, and usually is represented by earthquake induced acceleration, or PGA. Adversely, for buried pipelines, seismic wave propagation has less destructive effects on pipelines than permanent ground deformations, such as faulting, settlement, liquefaction induced lateral spreading, and landslide. Generally, combinations of both seismic effects are applied to pipeline, with different magnitudes and importance. Also, in most cases, wave propagation affects a larger area with small intensities, while PGDs have more devastating effects in limited locations. Gas pipelines are considered as an important part of lifelines, due to vast industrial and urban usage of gas and its increasing demand, as a clean and yet, cheap source of energy. Pipelined design is mainly based on mechanical and processing needs, where pipe size and thickness are determined. For highly seismic areas, structural design considering earthquake induced loads is also important. A very useful way to indicate wave propagation effects in pipelines, is Newmark’s theory, in which he relates pipeline axial strain caused by TGD effects, to the pick ground velocity, PGV, and apparent wave propagation velocity. After Newmark, different researchers have modified this relationship to account for different types of seismic waves, soil properties, pipe geometry and importance, and it has been mentioned in different codes and guidelines, such as ALA 2001, Eurocode 8, and ISDCOI-038. After calculation of axial strain, it shall be compared with allowable strain, which depends on pipeline material, and type of strain (tension or compression). Permanent ground deformations are other destructive seismic effects, and include different types, such as faulting, landslide, liquefaction induced lateral spreading, and settlement, for two former cases, codes such as IITK GSDMA have proposed some relationships. For landslide, direction of soil movement with regard to pipeline direction will cause axial, bending or combined forces in pipeline, which in turn initiates axial strains in pipeline section and shall be compared with allowable strains, as mentioned before. When faulting occurs, if fault trace crosses the pipeline, some deformations will happen in pipeline, which shall be calculated based in faulting mechanism, angle of crossing, soil properties, event magnitude, and pipeline importance, and subsequent strains shall be compared to allowable strains. To determine amount of ground deformations caused by faulting, various relationships are available, between which, those proposed by “Wells & Coppersmith” are more common. In these formula, pipeline deformation caused by faulting will be determined, multiplied by pipeline importance factors, and subsequent strains shall be compared with allowable strains. For aboveground pipelines, pipeline span or distance between supports is important and addressed in many technical documents and regulations. In some cases, such as trance Alaska oil pipeline, usage of flexible supports able to move horizontally in direction perpendicular to pipeline direction, has led to less seismic vulnerability in reported seismic events.

Concentric braced steel frames are commonly used in steel structures to resist the lateral forces especially in structures that are constructed in high seismic zones. Common assumption for the braces connection is pinned, but in practice, the connection of the braces may have partial rigidity, with regard to the details of the connection. Besides, the buckling load of the braces is affected by effective length factor that is affected by the conditions of the end connections. Therefore, changing the condition of the brace connection affects on the buckling load of brace and consequently on the behavior of the frame. In this paper, the behavior of two-dimensional inverted V-braced steel frames has been investigated with different number of stories (3, 5 and 10 stories) and bays (3 and 5 bays), which are different in connection conditions of braces. The lateral resistance of the frames is provided only by bracing system. The end connections of braces have been modeled in either pinned or rigid extreme states with proportional effective length factors. The frames were designed due to gravity and lateral loads according to the specifications of American Institute of Steel Construction (AISC), Allowable Stress Design (ASD), so that there was not any over-strength in the elements of the models. After designing, P Delta effects were considered and plastic hinges were assigned to the members to allow nonlinear behavior of frames according to the specifications of FEMA273. For this purpose, P plastic hinges were assigned to the braces with pinned connections, P-M plastic hinges were assigned to the columns and braces with rigid connections, and M plastic hinges were assigned to the beams. Initial static analyses of the frames were performed by the SAP2000 software. Then nonlinear static and nonlinear time-history analyses were done on the frames subjected to different strong ground motions to reveal the effects of the brace connections and effective length factor of the brace members on the seismic behavior of the steel frames. Triangular lateral loading pattern was applied to the frames for nonlinear static analyses. The results of the linear static analyses show that changing the end connections of the inverted V-braced frames has not considerable effects on the fundamental period, lateral stiffness and internal forces of the members of the frames. Moreover, the results of the nonlinear static analyses of the frames show that changing the brace connections does not affect on the yield point and shear capacity of the frames; but the effective length factor of the braces greatly affects on the behavior of the frames. The frames with lower effective length factors of braces have higher yield point and shear capacity than the other frames with greater effective length factors. Relative differences are about 8% to 67%. Similar results were obtained from nonlinear time-history analyses. According to these results, when the elements remain linear, differences in the response of models with different conditions for brace connections are negligible; whereas under some earthquakes that several plastic hinges were generated in the members, the differences in maximum base shear and roof displacement of the models are considerable.