نشریه مهندسی عمران و محیط زیست دانشگاه تبریز
سال چهل و دوم شماره 2 (پیاپی 67، تابستان 1391)

Topography magnification effects and the surroundings of structures had has an important role in their destruction after earthquakes. In this article, dynamic interaction of morning glory spillway of Barzu dam placed in Shirvan has been investigated in the frequency domain. Due to the position of structure, the earthquake occurrence is possible in the time when no water exists around the structure. Therefore, it is an important issue to consider the dynamic interaction between the spillway and its foundation. The solving method is based on boundary element method. First the formulation of boundary element method in the frequency domain is presented and then three examples are considered for verification of the results. The first example is a cantilever beam which is presented for verification of 3D elastodynamic finite medium problems. The second example is one layered valley which is presented for verification of one layered infinite medium. The third example is alluvial valley which is presented for verification of wave propagation in multi-layered infinite domains. After the verification, the foundation domain which is placed around some part of the structure has been modeled and analyzed separately, and then a model which has 2 sub domains has been presented. In order to have a realistic analysis, the problems have been modeled and analyzed three-dimensionally. In this study, the effect of some parameters such as shape irregularities, mechanical properties of materials (density, Poisson coefficient, shear modulus), stimulation frequency, incident wave type (P, SH, SV), azimuth and angle of incident wave on dynamic analysis result of the mentioned structure is investigated. The results of dynamic interaction analyses

Lead-rubber isolators have been successfully used so far in the footing of buildings to prevent transmission of earthquake excitations to the buildings. This isolator creates a layer in the building in which large horizontal deformations take place and less is transmitted to the construction. Although this layer reduces the vibrations considerably, however the costs and the maintenance and repair difficulties still remain in this system. Besides, isolator layers cannot be used in the tall buildings or in the buildings constructed over soft soil layers. In this research, it has been attempted to use lead-rubber layer as a passive energy absorbing device within the structure itself instead of the footings. The main aim of this study is the demonstration of the dynamic behavior of the lead-rubber isolator as a shear damper in the chevron bracing of structures. This device consists of several steel plates, rubber layers and a cylindrical lead core. The rubber layers are sandwiched between steel plates. In this paper, numerical analysis and experimental tests have been performed. First a three dimensional finite element model was created for the device and then it was analyzed using the Explicit Module of ABAQUS finite element program. The hyper-elastic properties were considered for the rubber layers. The results of this analysis were used to adjust the detail of the prototype damper properly. The device was manufactured at the Earthquake Engineering Laboratory of Urmia University and installed inside a SDOF steel frame. The frame was put on the shaking table of Urmia University and its responses under several earthquake excitations were recorded. By creating the numerical model of the SDOF frame in SAP2000 program and validating it with the experimental results, the effectiveness of the device in mitigation of the responses of multi-story frames was checked by conducting several nonlinear time history analyses on the structures. Three earthquake excitations were applied in these analyses. The results were indicating that great reductions in the stories drifts can be achieved by installation of the lead-rubber isolator in the chevron braces of structures as shear dampers. Another important advantage of this device is that it can be easily installed inside all kinds of structures.

Space structures are economical and aesthetic in appearance. They provide a unique solution to cover large column free areas. Sports stadia, industrial buildings, exhibition halls, cinema theatres, swimming pools, airport hangars, conference halls, etc, can be economically covered with the use of space structures. Space grids are characterized as two-way or three-way, depending on whether the members intersecting at a node run in two or three directions. Three-way grids are very extremely strong and lead to a very uniform stress distribution. Barrel vault is one of the oldest architectural forms, used since antiquity. The popularity of barrel vaults derived partly from the economy of these structures, as all arches could be constructed as identical members. Braced barrel vaults are developable surfaces of zero Gauss curvature generated by the movement of a curve, known as the. directrix., over a generator straight line. The directrix may be a circular arc, an ellipse, a catenary, a parabola or a cycloid. Single-layer barrel vaults show a range of different instability modes, namely, member instability, node instability, line instability and overall instability, normally each occurring under different loading and boundary conditions. There are two important types of instability, namely: limit point instability and bifurcation point instability. In limit point instability, a structure with nonlinear un-stiffening characteristics loses its stability by a sudden buckling into a mode of deformation which is in accordance with the initial mode of deformation. However, in the bifurcation point instability, a structure with nonlinear softening characteristics may lose its stability by a sudden buckling into a mode of deformation which is quite distinct from the initial unstiffening mode. There are three types of bifurcation point instability, namely: asymmetric bifurcation, stablesymmetric bifurcation and unstable-symmetric bifurcation. However, in most cases, space structures show limit point instability and unstable-symmetric bifurcation point instability. In single-layer braced barrel vaults, the effect of geometric nonlinearity is very important. In fact, due to small angles between the members, deflections and member forces are large and thus significant changes in structural geometry occur. Also, in these structures the possibility of plasticity occurring in the members causes a reduction in their load carrying capacity. Plasticity can occur before the first limit point load in single-layer braced barrel vaults, so despite the fact that the main source of nonlinearity in these structures is geometric, the inclusion of plasticity in the analysis is necessary to safely predict their true collapse load and post-collapse behavior. Therefore, geometric and material nonlinearities should be considered in the analysis of these structures. In the present study, the stability of singlelayer barrel vaults with three-way triangular configuration is investigated by carrying out nonlinear collapse analysis using finite element method. The effects of different parameters, such as initial geometric imperfections, load on a single node, loads on some nodes, unsymmetrical load distribution with different intensities, length to span ratio, height to span ratio, boundary conditions and material yield stress are considered on the stability behavior of single-layer barrel vaults. According to static collapse analyses of studied models, three collapse mechanisms were determined as overall collapse, local collapse with snap-through and localcollapse without snap-through. In the case of local collapse with snap-through, it has been indicated that to evaluate realistic response of the structure, the static collapse analysis is not adequate by itself, and a dynamic snap-through analysis must be performed, as well.

One of the most dangerous kinds of the damage in concrete structural members is shear failure with a sudden and brittle failure. Hence, according to numerous codes of practice, using sufficient transverse reinforcement in the critical region is necessary. For this reason, to improve the shear behavior and change the brittle behavior to ductile one, steel fiber can play an important role. In this research, the effect of short wavy steel fiber in the shear behavior of reinforced concrete beams with normal strength was investigated. The experiments were conducted in three groups. in the first group, concrete containing steel fiber without stirrup (BF), in second group, concrete containing stirrup without steel fiber (BS) and in the third group, concrete containing steel fiber with stirrup (BFS) have been placed. The results indicated that by replacement of the fiber with the same volume instead of stirrup higher stiffness and ductility can be reached without the reduction in strength. Also, increasing steel fiber, the shear failure will be changed into bending failure indicating rapid growth of the shear strength in comparison with bending strength. Distances and the length of cracks in the beams containing steel fiber were shorter. These cracks were created discontinuously and after continuing the loading, were connected to each other. Moreover, these cracks were along with a lower voice in comparison with beams without fiber during the loading, indicating the more uniform redistribution of stresses. Finally, can be said, for the prevention of the sudden cracks and the better redistribution of stresses, the increase of the amount of the steel fiber is more effective than the increase of the amount of stirrup.

Modeling of the soil behavior subjected to monotonic loading is of great importance. In traditional old models, incapability of modeling sand strength and subsequently calculation errors lead to unreal definition of sand strength. Furthermore, these models define samples of a soil with different densities and mean stresses as different soils and there is no comprehensive control on state changes during the loading. On the other hand, more advanced models are complicated caasing some problems in their usage. Thus, the necessity of a noble plain model which allows quick and applicable calculations is apparent. In this article sand behavior subjected to monotonic undrained loading is defined from loose state to dense state. In general all stress paths of monotonic loading of a sand in total ranges of confining stress and density throughout confining stress-shear stress space pq and in axial strain-confining stress space å1-p are comparable. Consequently monotonic sand behavior under shear condition can be defined in a new space by means of some state parameters. For this purpose, two new state parameters are presented as: confining stress ratio Rp=p/p0 and shear stress ratio Rs=q/qa. Moreover state pressure index Ip=p/pc is benefited in which p is current confining stress, p0 is initial confining stress, q is current shear stress and pc is critical state confining stress. Due to constant void ratio in undrained test, shear strength of sand is a function of confining stress. Thus by selecting an arbitrary amount of confining stress ratio in stress path prior to phase transformation point, Rpa, corresponding shear stress of Rpa is determined for all stress paths. This shear stress of each stress path is called qa. To characterize the effect of density and density dependent effect of confining stress on shear strength, two spaces å1-Rp and Rp-Rs are considered. Stress paths vary by the flux in initial state pressure index IP0 which lead to form a group of stress paths. In å1-Rp spaces, stress-strain paths are capable of being divided into two distinct groups of paths with 1) IP0≥1 and 2) IP0<1. In Rp-Rs spaces, stress paths with a same void ratio are in a same envelope. Each envelope of stress path is defined as a function of void ratio. In order to investigate effect of IP0 on sand behavior during strain increase, in a quantitative form, diagrams of IP0-Rp are introduced. A diagram of IP0-Rp is defined by determining IP0 of each stress path, in a same level of strain, and diagram of IP0-Rp is achieved for that level of strain. Other diagrams are determined using the same way. Diagrams of Rp-Rs are appropriately capable of indicating the effect of IP0 on sand, and the twoimportant factors controlling their variation are IP0 and void ratio. Subsequently, a relation between confining stress ratio, strain and IP0 and a relation between shear stress ratio, confining stress ratio and void ratio are developed in the form of a model. A simulation of the developed model is performed and employed to investigate the behavior of two different types of sands using two different types of loadings, stress controlled and strain controlled. The obtained results of the model are in good agreements with the available valid laboratory results, which show its capability in defining sand behavior under undrained condition. This model provides a definition of a sand behavior by some specified constants and softening and hardening behavior, and dilative and contractive soil response during loading process, is described in all ranges of density and stresses within the framework of this model.

Prediction of settlement and its distribution for shallow foundation which are among the most important problems for geotechnical engineers are influenced by various factors such as shape and stiffness of foundation, geometry, weight and stiffness of structure, type, stiffness, stress-strain history and thickness of soil layers. In the literature, there are different classical relationships for prediction of foundation settlement. These relationships are generally based on elasticity theory for immediate settlement and oedometric conditions for time-dependent settlement. Due to the limitation in consideration of material behavior, geometry and boundary conditions and loading, related to these relationships, application of calculation codes with proper constitutive model can be regarded as a satisfactory method. In this paper, a set of finite element analyses using an elastoplastic soil model have been performed for prediction of shallow foundation settlement and its distribution. Distinguished aspect of these analyses is that the soil constitutive model has been calibrated using Plate Load Test results. Similarity of stress-strain field for the Plate Load Test and real foundation, on one part, and the proper relationship between elastic and hardening modulus of constitutive model and confining stress, on the other part, led to enriching the results of prediction qualitatively and quantitatively. Using the calculation code (Plaxis 3D) calibrated based on Plate Load Test, the effects of foundation width and depth on settlement and its distribution were studied. Besides, classical relationships of settlement estimation were evaluated by comparison with numerical results. The obtained results indicate that, firstly, the dependence of foundation settlement to width is significantly nonlinear, so existing classical relationships don’t consider this important aspect. Settlement distribution depends on soil and footing stiffness. In the case of granular soils due to significant dependence of elastic and plastic modulus to confining stress, settlement in foundation edge is more than center. The results of analyses also indicated that settlement and its distribution depend nonlinearly on foundation depth. Based on settlement distribution of foundation, it was found that sub grade reaction modulus is not a constant parameter for a given problem. On the other hand, settlement distribution dictates the sub grade reaction modulus distribution for foundation analysis and design.

Up-lift force is one of the efficient parameters in loading over different types of hydraulic structures. In order to determine up-lift force, seepage analysis is required. This force is usually computed in laminar seepage, where the Darci law is the governing equation. Methods of seepage analysis can be classified in three general categories: analytical, experimental and numerical. If Darci conditions are not appointed in media, the governing seepage equation in soil will be nonlinear and it does not any analytical solution, so numerical or experimental methods must be applied to solve the problem. This paper explores up-lift force through seepage analysis in non- Dacian flows. In general, researchers have issued two different categories for modeling of non Darcian flows. The first category considers the relationship between flow velocity and hydraulic gradient and the second one deal with equations between Darcy-Weisbach friction coefficient and Reynolds number. In this study, equations of first category between flow velocity and hydraulic gradient have been combined with flow continuity equation for non-Dacian seepage modeling through numerical methods. Recently, the discrete singular convolution (DSC) has been proposed and applied to many practical problems. The objective of this paper is to examine the DSC algorithm for solving the mentioned equations and to compare the DSC with finite volume (FV) method, whose advantage for seepage analysis has been documented. Comprehensive comparison study between DSC approach and well-known method of FV is obtained for estimating up-lift force over the foundation of a concrete dam. Results are compared with common Bligh method. Results of two numerical methods agree very well. Thus, DSC can be applied for nonlinear equations governing the seepage in porous media to calculate the up-lift force in non-Darcian seepage conditions. Furthermore, because of consideration of various parameters such as diameter of particles, different coefficients of non-Darcian seepage, it is observed that the DSC and FV approaches performs practically and computationally much better for this study rather than Bligh method. Moreover it is concluded that, although Bligh method was applied by designers without considering the flow regime, results indicate that DSC can be utilized for designing purposes when the sheet pile below the dam body is locating on downstream. However when the sheet pile below the dam body is located upstream, this method can not be used because the calculated up-lift force is not on the safe side.