Date

2015

Author

Etemadi, Ali

Metadata

Show full item record

Item Usage Stats

198
views

114
downloads

Cite This

In assessing the damage originating from strong ground motions in building frames, it is necessary to identify properly the post-yield hysteresis degrading behavior of structural components that are well correlated with structural response and in turn, with damage. Likewise, structural damage during the ground motion is due to excessive interstory drift ratio; hence more realistic estimation of interstory drift demands has a significant role in the seismic evaluation of frame buildings. Existing approaches used to calculate the drift spectrum are valid in elastic ranges and cannot count for overestimated drift demands due to the post-yielding behavior of structural systems. A simple procedure to estimate the spectra of maximum interstory drift demands in shear-type frames that respond in post-elastic limits is described in this thesis, and the effect of hysteresis deterioration properties on seismic demands is clarified. Afterwards, the modification factors are proposed to incorporate the hysteresis degradation effects parametrically. These factors are defined with respect to the corresponding elastic drift demands. The closed-form drift spectrum is adopted as reference spectrum to validate the proposed spectrum ordinates in elastic ranges. The closed-form drift spectrum is derived based on the continuous shear-beam model and wave propagation theory. To derive the drift spectrum though the proposed method, series of simple shear frames are designed to be consistent with the continuous shear-beam models thereafter by systematic variation of their column stiffness and story mass properties, a reasonable period range (0.3-2.4 s) is obtained. When these frames are subjected to the ground motions considered, their dynamic responses are computed by time history response analysis method, and this is performed in both elastic and inelastic ranges. A smooth hysteretic model is adopted to incorporate the nonlinearity of structural members into the nonlinear time history calculations by meaningful magnitude variations of the control parameters. In this way the different types of hysteresis degrading properties (i.e. stiffness decay, strength deterioration and hystereis pinching) are modeled by considering the rotational springs at both ends of each column for all stories. This model is extended based on the classical differential Bouc-Wen model. Access to experimental results of the cyclic force-deformation characteristics of components typical to the structure being analyzed provides the best means of specifying the above degrading parameters. The parametric identification study is carried out to clarify the relationship between control parameters and the response hysteresis loops as well as provide a comprehensive range of the basic parameters. This is used to formulate a model for the desired hysteresis loop where all its parameters are physically meaningful. It is assumed that these quantities are collected from the cyclic loading tests, which reflects the realistic hysteresis decay characteristics of RC columns when exposed to severe cyclic loadings. The drift demands gained through the wave propagation solution, and the results of 105,000 response history analyses in both elastic and inelastic ranges are used for calculating the error statistics. Such massive numbers of repeated processes are very exhausting to conduct manually, and this is the underlying factor in the development of a code to perform these time-consuming processes automatically. The least squared regression analysis is conducted on the intact differences between both elastic and inelastic spectrum ordinates to get the smooth variation functions. The modification coefficients are proposed as the function of vibration period and some other system dependent variables. In the dataset of records, a substantial number of near-fault ground motions are involved, which make large seismic demands to the structures. Such a number of selected records are an indication of the reliability of statistical results. This method could apply to the rapid seismic evaluation of the existing poor-detailed, non-ductile buildings.

Subject Keywords

Structural analysis (Engineering)., Buildings, Structural frames., Shear (Mechanics)., Statics.

URI

http://etd.lib.metu.edu.tr/upload/12619300/index.pdf
https://hdl.handle.net/11511/25231

Collections

Graduate School of Natural and Applied Sciences, Thesis