Zaman integrasyon şemalarının eşdeğer dinamik test sonuçlarına etkisi

SEZER ÖZTÜRK

Zaman integrasyon şemalarının eşdeğer dinamik test sonuçlarına etkisi

Thesis Type: Postgraduate

Institution Of The Thesis: Yıldız Technical University, Graduate School of Natural and Applied Sciences, İNŞAAT MÜHENDİSLİĞİ ANABİLİM DALI, Turkey

Approval Date: 2017

Thesis Language: Turkish

Student: SEZER ÖZTÜRK

Supervisor: Bilge Doran

Open Archive Collection: AVESIS Open Access Collection

Abstract:

Determination of the behavior of structures and structural elements under the influence of different loads is a topic that engineers and researchers have been interested in for many years. Different methods are used to examine the performance of structures, especially after earthquakes. These are: i) site surveys after earthquakes, ii) mathematical models, and iii) laboratory tests. Due to the fact that the pseudo dynamic test method is more economical and easier to implement than the shake table tests, it has become a reliable alternative and frequently used because it deals with building capacity and earthquake demand. As a basic principle, the structure is examined step by step with integration method. The simulation is carried out at non-real time so that damage formation and development can be observed on the sample. The reaction forces obtained from the physical test of the structure at each time step are mathematically combined with the inertia and damping effects of the structure in a computer model and the displacement values desired from the structure at the next time step of the earthquake are determined. The test continues with the application of these displacement values to the physical structure. In situations where the loading is randomly changing or the system exhibits nonlinear behavior, time integration scheme algorithms are used to generate a solution to the dynamic equation of the motion. In this study, the effects of the Central Differences Method and the α-OS Method on the dynamic behavior, which are used in the pseudo dynamic test method, are investigated. The dynamic behavior of a single degree of freedom system has been studied both analytically and experimentally. In these investigations, the dynamic load is used as the instantaneous constant load and the selected earthquake acceleration records. Numerical analyzes have been carried out in both the elastic and inelastic regions by applying constant load and ground acceleration records to the system with both integration methods. In these analyzes, the effect of the two integration methods and different time step interval values to the pseudo dynamic analyzes was investigated. Experimental studies have been carried out at Fatih Sultan Mehmet Vakıf University under "Structural Mechanics and Earthquake Research Laboratory" under constant load, at different time step intervals, using α-OS method and compared with numerical analysis results. In the analytical results obtained for the very small values of the time step interval, it is seen that the solutions of both integration methods give very close results. However, as the time step interval is increased, the stability has deteriorated in the solutions. In the Central Difference Method solutions, the period of the structural system has decreased whereas the period of the structural system has been increased in the solutions of α-OS Method. In experimental studies, results were close to numerical analysis results. It has been seen that the pseudo dynamic experimental infrastructure gave realistic results in the studies carried out. The effects of numerical integration methods and time step interval values on the analysis are examined. It is thought that the study may be able to generate ideas in the selection of the appropriate time step interval to be used in pseudo dynamic experiments.