Published Online: 1 March 2012
Page Count: 8
Gundes Bakir, Pelin
Professor, Dept. of Civil Engineering, Istanbul Technical Univ., Maslak, Istanbul
Eksioglu, Ender M.
Assistant Professor, Dept. of Electronics and Communications Engineering, Istanbul Technical Univ., Maslak, Istanbul
Dept. of Civil Engineering, Istanbul Technical Univ., Maslak, Istanbul
(Received 28 March 2011; accepted 10 January 2012)
The complex mode indication function (CMIF) technique is a singular value decomposition enhancement of the classical peak picking (PP) technique. The disadvantages of the PP technique, namely, its inability to identify closely spaced modes, are largely resolved in CMIF due to the singular value decomposition. Recently, a frequency domain based technique that uses the biased frequency response functions obtained from the Hilbert transform of power spectral densities has been proposed. In this study, the PP, CMIF, and Hilbert transform techniques are successfully applied to the measured output data from a typical instrumented school building in Istanbul. The study clearly shows that the techniques can successfully identify the modal parameters and give similar results for the eigenfrequencies and mode shapes. However, there are some differences in the damping ratio estimates.
Paper ID: JTE103862