Stochastic response of suspension bridges for various spatial variability models

Ahmet C. Altunışık,Süleyman Adanur,Kurtuluş Soyluk,A. Aydın Dumanoğlu 국제구조공학회 2016 Steel and Composite Structures, An International J Vol.22 No.5

The purpose of this paper is to compare the structural responses obtained from the stochastic analysis of a suspension bridge subjected to uniform and partially correlated seismic ground motions, using different spatial correlation functions commonly used in the earthquake engineering. The spatial correlation function employed in this study consists of a term that characterizes the loss of coherency. To account for the spatial variability of ground motions, the widely used four loss of coherency models in the literature has been taken into account in this study. Because each of these models has its own characteristics, it is intended to determine the sensitivity of a suspension bridge due to these losses of coherency models which represent the spatial variability of ground motions. Bosporus Suspension Bridge connects Europe to Asia in Istanbul is selected as a numerical example. The bridge has steel towers that are flexible, inclined hangers and a steel box-deck of 1074 m main span, with side spans of 231 and 255 m on the European and Asian sides, respectively. For the ground motion the filtered white noise model is considered and applied in the vertical direction, the intensity parameter of this model is obtained by using the S16E component of Pacoima Dam record of 1971 San Fernando earthquake. An analytically simple model called as filtered white noise ground motion model is chosen to represent the earthquake ground motion. When compared with the uniform ground motion case, the results obtained from the spatial variability models with partial correlation outline the necessity to include the spatial variability of ground motions in the stochastic dynamic analysis of suspension bridges. It is observed that while the largest response values are obtained for the model proposed by Harichandran and Vanmarcke, the model proposed by Uscinski produces the smallest responses among the considered partially correlated ground motion models. The response values obtained from the uniform ground motion case are usually smaller than those of the responses obtained from the partially correlated ground motion cases. While the response values at the flexible parts of the bridge are totally dominated by the dynamic component, the pseudo-static component also has significant contributions for the response values at the rigid parts of the bridge. The results also show the consistency of the spatial variability models, which have different characteristics, considered in this study.

Collapse of steel cantilever roof of tribune induced by snow loads

Ahmet C. Altunışık,Şevket Ateş,Metin Hüsem,Ali F. Genç 국제구조공학회 2017 Steel and Composite Structures, An International J Vol.23 No.3

In this paper, it is aimed to present a detail investigation related to structural behavior of laterally unrestrained steel cantilever roof of tribune with slender cross section. The structure is located in Tutak town in A&$287;rı and collapsed on October 25, 2015 at eastern part of Turkey is considered as a case study. This mild sloped roof structure was built from a variable I beam, and supported on steel columns of 5.5 m height covering totally 240 m² closed area in plan. The roof of tribune collapsed completely without any indication during first snowfall after construction at midnight a winter day, fortunately before the opening hours. The meteorological records and observations of local persons are combined together to estimate the intensity of snow load in the region and it is compared with the code specified values. Also, the wide/thickness and height/thickness ratios for flange and web are evaluated according to the design codes. Three dimensional finite element model of the existing steel tribune roof is generated considering project drawings and site investigations using commercially available software ANSYS. The displacements, principal stresses and strains along to the cantilever length and column height are given as contour diagrams and graph format. In addition to site investigation, the numerical and analytical works conducted in this study indicate that the unequivocal reasons of the collapse are overloading action of snow load intensity, some mistakes made in the design of steel cantilever beams, insufficient strength and rigidity of the main structural elements, and construction workmanship errors.

Structural response relationship between scaled and prototype concrete load bearing systems using similarity requirements

Ahmet C. Altunışık,Ebru Kalkan,Hasan B. Başağa 사단법인 한국계산역학회 2018 Computers and Concrete, An International Journal Vol.21 No.4

This study is focused on the investigation for similitude the requirements between prototype and scaled models to determine the structural behavior of concrete load bearing systems. The scaling concept has been utilized in many engineering branches, has been assisted to engineers and scientists for obtain the behavior of the prototype by using scaled model. The scaling can be done for two purposes, either scaling up or scaling down depending upon the application. Because, scaled down models are the experimentation on scaled models is cheaper than huge structures. These models also provide facilities for experimental work. Similarity relationships between systems are created either by field equations of the system or by dimensional analysis. Within this study, similarity relationships were obtained by both methods. The similarity relations obtained are applied to different load bearing systems and it is determined that the similarity relation is a general expression. In this study, as an example, column, frame, cantilever beam and simple beam are chosen and 1/2, 1/5 and 1/10 scales are applied. The results are compared with the analytical results which are obtained by creating of the finite element models with SAP2000 software of different scaled load bearing systems. The analysis results of all systems are examined and it is determined that the scale factors are constant depending on the scale types for different load bearing systems.

Investigation of earthquake angle effect on the seismic performance of steel bridges

Ebru Kalkan,Ahmet C. Altunışık 국제구조공학회 2016 Steel and Composite Structures, An International J Vol.22 No.4

In this paper, it is aimed to evaluate the earthquake angle influence on the seismic performance of steel highway bridges. Upper-deck steel highway bridge, which has arch type load bearing system with a total length of 216 m, has been selected as an application and analyzed using finite element methods. The bridge is subjected to 1992 Erzincan earthquake ground motion components in nineteen directions whose values range between 0 to 90 degrees, with an increment of 5 degrees. The seismic weight is calculated using full dead load plus 30% of live load. The variation of maximum displacements in each directions and internal forces such as axial forces, shear forces and bending moments for bridge arch and deck are attained to determine the earthquake angle influence on the seismic performance. The results show that angle of seismic input motion considerably influences the response of the bridge. It is seen that maximum arch displacements are obtained at X, Y and Z direction for 0°, 65° and 5°, respectively. The results are changed considerably with the different earthquake angle. The maximum differences are calculated as 57.06%, 114.4% and 55.71% for X, Y and Z directions, respectively. The maximum axial forces, shear forces and bending moments are obtained for bridge arch at 90°, 5° and 0°, respectively. The maximum differences are calculated as 49.12%, 37.37% and 51.50%, respectively. The maximum shear forces and bending moments are obtained for bridge deck at 0°. The maximum differences are calculated as 49.67%, and 49.15%, respectively. It is seen from the study that the variation of earthquake angle effect the structural performance of highway bridges considerably. But, there is not any specific earthquake angle of incidence for each structures or members which increases the value of internal forces of all structural members together. Each member gets its maximum value of in a specific angle of incidence.

Construction stages analyses using time dependent material properties of concrete arch dams

Barış Sevim,Ahmet C. Altunışık,Alemdar Bayraktar 사단법인 한국계산역학회 2014 Computers and Concrete, An International Journal Vol.14 No.5

This paper presents the effects of the construction stages using time dependent material properties on the structural behaviour of concrete arch dams. For this purpose, a double curvature Type-5 arch dam suggested in ‘‘Arch Dams’’ symposium in England in 1968 is selected as a numerical example. Finite element models of Type-5 arch dam are modelled using SAP2000 program. Geometric nonlinearity is taken into consideration in the construction stage analysis using P-Delta plus large displacement criterion. In addition, the time dependent material strength variations and geometric variations are included in the analysis. Elasticity modulus, creep and shrinkage are computed for different stages of the construction process. In the construction stage analyses, a total of 64 construction stages are included. Each stage has generally 6000 m3 concrete volume. Total duration is taken into account as 1280 days. Maximum total step and maximum iteration for each step are selected as 200 and 50, respectively. The structural behaviour of the arch dam at different construction stages has been examined. Two different finite element analyses cases are performed. In the first case, construction stages using time dependent material properties are considered. In the second case, only linear static analysis (not considered construction stages) is taken into account. Variation of the displacements and stresses are obtained from the both analyses. It is highlighted that construction stage analysis using time dependent material strength variations and geometric variations has an important effect on the structural behaviour of arch dams. The maximum longitudinal, transverse and vertical displacements obtained from construction stages and static analyses are 1.35 mm and 0 mm; -8.44 and 6.68 mm; -4.00 and -9.90 mm, respectively. In addition, vertical displacements increase from the base to crest of the dam for both analyses. The maximum S11, S22 and S33 stresses are obtained as 1.60MPa and 2.84MPa; 1.39MPa and 2.43MPa; 0.60MPa and 0.50MPa, respectively. The differences between maximum longitudinal, transverse, and vertical stresses obtained from construction stage and static analyses are 78%, 75%, and %17, respectively. On the other hand, there is averagely 12% difference between minimum stresses for all three directions.