SYSTEM IDENTIFICATION FOR 4 TYPES OF STRUCTURES IN ISTANBUL BY FREQUENCY DOMAIN DECOMPOSITION AND STOCHASTIC SUBSPACE IDENTIFICATION METHODS

Dominik Stengel

(Thesis Supervisor: Eser Durukal)

ABSTRACT

Two rather new techniques in the field of system identification, called frequency domain decomposition FDD, that can be extended to enhanced frequency domain decomposition EFDD and stochastic subspace identification SSI are presented and explained in detail.

Those methods arc used to conduct an operational modal analysis OMA of Fatih Sultan Mehmet Bridge under strong wind excitation. Natural frequencies, mode shapes and damping ratios arc obtained and discussed. Two other sets of data arc analysed in the same way but best results are achieved under wind excitation .

Further, ambient vibration test on Hagia Irene and Mihrimah Sultan minaret are described and analysed. Again natural frequencies, mode shapes and damping ratios arc calculated and discussed. Both methods proved to be helpful tools in fast analysis of those tests.

Last, incoming data from the monitoring system of Hagia Sophia is used to conduct system identification and investigate the atypical seismic response behaviour of the structure, like first noticed in 1999 Kocaeli earthquake. The characteristic drop of first two eigenfrequcncies is attested during a minor earthquake occurred in Klitahya on 1ih February 2009 even though much smaller in value.