Ongoing study on a scaled steel framed structure in a shaking table: preliminary results on the relation between fundamental period variations and damage

Abstract

The feasibility of the use of fundamental period variations in buildings as damage indicators is analysed by means of a scaled steel framed structure mounted on a lab shaking table. The structure is a five-floor rectangular build with one and two frames in the transversal and longitudinal directions. Beams are steel bars with 20x10 mm sections, while the sections of columns are semi-squares whose two sides have 20x2 mm. The dimensions of the structure are 1440, 770 and 1700 mm, in the longitudinal, transversal and vertical directions, respectively. The mass of the floors is simulated by lead blocks, weighting 4.5 kg each, and located in the middle of the beams between columns, so that 157.5 kg (5 floors x 7 frames x 4.5 kg) are added to the mass of beams and columns. Beams and columns are joined by bolts, which can be tightened or loosened to simulate different degrees of damage. The structure is instrumented with six 2D accelerometers, recording the longitudinal and transversal motions of the response of the structure to specific impulsive excitations, simulating seismic actions. Two data sets, Data_1 and Data_2, corresponding to the sound and slightly damaged structure, are analysed. Signals where carefully analysed in the time and frequency domains. Power Spectral Density allowed us to detect four clear peaks at frequencies of 2.71, 3.78, 7.64 and 9.63 in the DATA_1 data set. The simulated damage, which is considered slight-to-moderate damage, induces decreases of about 8.6%, 7%, 10% and 4.5% for the four frequencies detected.

Postprint (published version)