| The problem of modal identification in civil structures is of crucial importance, and thus has
been receiving increasing attention in recent years. Vibration-based methods are quite
promising as they are capable of identifying the structure’s global characteristics, they are
relatively easy to implement and they tend to be time effective and less expensive than most
alternatives [1]. This paper focuses on the off-line structural/modal identification of civil
(concrete) structures subjected to low-level earthquake excitations, under which, they remain
within their linear operating regime. Earthquakes and their details are recorded and
provided by the seismological network of Crete [2], which “monitors” the broad
region of south Hellenic arc, an active seismic region which functions as a natural
laboratory for earthquake engineering of this kind. A sufficient number of seismic events
are analyzed in order to reveal the modal characteristics of the structures under
study, that consist of the two concrete buildings of the School of Applied Sciences,
Technological Education Institute of Crete, located in Chania, Crete, Hellas. Both
buildings are equipped with high-sensitivity and accuracy seismographs – providing
acceleration measurements – established at the basement (structure’s foundation) presently
considered as the ground’s acceleration (excitation) and at all levels (ground floor, 1st
floor, 2nd floor and terrace). Further details regarding the instrumentation setup and
data acquisition may be found in [3]. The present study invokes stochastic, both
non-parametric (frequency-based) and parametric methods for structural/modal
identification (natural frequencies and/or damping ratios). Non-parametric methods
include Welch-based spectrum and Frequency response Function (FrF) estimation,
while parametric methods, include AutoRegressive (AR), AutoRegressive with
eXogeneous input (ARX) and Autoregressive Moving-Average with eXogeneous
input (ARMAX) models[4, 5]. Preliminary results indicate that parametric methods
are capable of sufficiently providing the structural/modal characteristics such as
natural frequencies and damping ratios. The study also aims – at a further level of
investigation – to provide a reliable statistically-based methodology for structural health
monitoring after major seismic events which potentially cause harming consequences in
structures.
Acknowledgments
This work was supported by the State Scholarships Foundation of Hellas.
References
[1] J. S. Sakellariou and S. D. Fassois, "Stochastic output error vibration-based damage
detection and assessment in structures under earthquake excitation," Journal of Sound and
Vibration, vol. 297, pp. 1048-1067, 2006.
[2] G. Hloupis, I. Papadopoulos, J. P. Makris, and F. Vallianatos, "The South
Aegean seismological network – HSNC," Adv. Geosci., vol. 34, pp. 15-21,
2013.
[3] F. P. Pentaris, J. Stonham, and J. P. Makris, "A review of the state-of-the-art of
wireless SHM systems and an experimental set-up towards an improved design," presented at
the EUROCON, 2013 IEEE, Zagreb, 2013.
[4] S. D. Fassois, "Parametric Identification of Vibrating Structures," in Encyclopedia of
Vibration, S. G. Braun, D. J. Ewins, and S. S. Rao, Eds., ed London: Academic Press,
London, 2001.
[5] S. D. Fassois and J. S. Sakellariou, "Time-series methods for fault detection and
identification in vibrating structures," Philosophical Transactions of the Royal Society A:
Mathematical, Physical and Engineering Sciences, vol. 365, pp. 411-448, February 15 2007. |