Kuvvetli yer hareketi seçiminin sismik şev deplasmanı üzerindeki etkisinin modellenmesi.

Assessing the earthquake performance of a slope and estimating the seismically-induced slope displacements is one of the most complicated tasks in geotechnical earthquake engineering. The source of the complication includes: i) defining the soil properties and their variability within the limits of the available geological/geotechnical information, and ii) executing a proper ground motion selection and scaling procedure for the dynamic numerical analysis, which are generally limited in number for most of the engineering applications. The objective of this study is to model the uncertainty due to ground motion selection in the seismically-induced soil slope displacements estimated by the dynamic numerical analysis. For this purpose, fifty horizontal pairs of ground motions recorded at rock stations from the Pacific Earthquake Engineering Research Center database are selected and scaled up to 1.0g of maximum horizontal acceleration to create the candidate ground motion dataset of 300 recordings. These recordings are utilized in PLAXIS-2D software to perform the dynamic numerical analysis for eight different slopes in dry (drained) soil. Analysis results showed that the standard deviation of a simplified prediction model for seismically-induced permanent slope displacements lie within the range of 0.28-0.36 (log units) and does not show a significant variation with factor of safety in static conditions or angle of the slope geometry. At the end of the study, 18 ground motion components that represent the median behaviour are selected and provided for the future studies. These recordings may be preferred in the numerical dynamic analysis to examine slope displacements and when combined with the estimated standard deviations, the body and the range in seismic demand models for earthquake-induced slope displacements may be properly modelled.Thesis (M.S.) -- Graduate School of Natural and Applied Sciences. Civil Engineering