Damage detection of a steel–concrete composite frame by a multilevel approach: Experimental measurements and modal identification

Abstract

Structural Health Monitoring (SHM) and Experimental Modal Analysis (EMA) have revealed to be really efficient tools for the damage detection of structures and structural elements in many engineering fields (Shock Vib. Dig. 1998; 30(2):91–105; Philos. Trans. R. Soc. A 2007; 365(1851):303–315; Mech. Syst. Signal Process. 2003; 17(1):133–142; Philos. Trans. R. Soc.: Math. Phys. Eng. Sci. 2001; 359(1778): 131–149; Struct. Control Health Monit. 2006; 14:1083–1100; Mech. Syst. Signal Process. 2004; 17(1): 83–89). Unfortunately the great variety of morphologies, construction materials and structural schemes makes these techniques not easily applicable to buildings, due to inherent problems related to the location and the extent of damage, the sensitivity of the dynamic response to damage, the choice of damage indexes to be used, the sensors' location, etc. On the other hand, the modern seismic capacity design of buildings in earthquake‐prone areas tends to locate dissipative zones in well‐determined portions of structures (CEN, European Committee for Standardization, EN 1998‐1. Eurocode 8: Design of Structures for Earthquake Resistance. Part 1: General Rules, Seismic Actions and Rules for Buildings, European Community, Brussels, Belgium, 2005); thus, the application of vibration‐based damage detection techniques to earthquake‐resistant structures seems to be very promising even if nowadays there are only few studies on these arguments. The present paper reports the experiences of a vibration‐based damage identification study applied to a steel–concrete composite frame structure, suitably designed to be high ductile according to Eurocode 8 and localizing the seismic energy dissipation in the beam‐to‐column joints. The structure was subjected to series of pseudo‐dynamic (PsD) and cyclic tests with increasing peak ground acceleration at the European Laboratory for Structural Assessment of Joint Research Centre at Ispra (VA, Italy). The damaging phenomena, caused by PsD tests, were assessed and quantified by means of a multi‐level vibration‐based approach suitably designed in order to evaluate the changes in the global dynamic structural response and to experimentally estimate the reduction in joint stiffness. Copyright © 2008 John Wiley & Sons, Ltd.

Damage analysis of a steel–concrete composite frame by finite element model updating

“…In fact, comparing the stiffness values in Section 3 and Tables 5, 8 and 11, it is possible to observe a maximum difference of 12% for the undamaged state, while for ULS and CLS the difference grows rapidly. These stiffness values were also in sufficient agreement with those obtained by the damage assessment procedure based on EMA local results applied on an exterior and interior joint [37] and with the values obtained by the cyclic tests on substructures at University of Pisa [35], listed in the Table 15. The two Local Layouts B and C (cited in Table 15) refer to an interior and an exterior sensor layout measurement from which, once modal identification is applied, it has been possible to evaluate, by means of the methodology described in [37], the stiffness of the interior and exterior beam-to-column connections and column shear panel.…”

“…Due to the peculiarity of seismic behavior of the analyzed structure, the application of a EMA allowed one to recognize progressive modifications of the global structural response and to measure and assess local damage in the beam-to-column joints devoted to energy dissipation, as stated in the adopted seismic capacity design. A complete description of the experimental results in terms of identified frequencies, damping ratios, mode shapes and joint stiffness is reported in [37]. tests were repeated at three damage levels in order to assess and quantify the progressive deterioration of the joints due to PsD tests.…”

“…The eigenfrequency residuals were complemented with spatial information about the dynamic behavior of the structure, including mode shape displacement differences in the residual vector. In the present case the mode shapes were identified by an Output-Only Modal Analysis procedure [37] and they are not absolutely scaled; so their residuals were formulated in a relative form:…”

Seismic performance of a 3D full‐scale high‐ductile steel–concrete composite moment‐resisting frame—Part II: Test results and analytical validation

“…This fact implies an increasing of the structural overstrength. Moreover, that period lengthened significantly as a result of the damage, damping the inertial forces [26][27][28]. Equivalent viscous damping ratios not exceeding 1% are recommended for non-linear dynamic analysis of the bare frame system without non-structural components.…”

Damage identification of a 3D full scale steel–concrete composite structure with partial‐strength joints at different pseudo‐dynamic load levels

“…Both studies were successful in joint identification, although only simple beam-to-column joint models in the elastic range were considered. Chellini et al [16] used modal identification in order to perform damage detection of semi-rigid partial-strength joints experiencing severe inelastic damage in the composite structure analysed in [5,6] and characterized by low masses, low damping and complex geometry. However, the overall understanding of damage 1221 evaluation in these structures and the issues related to the application of model updating techniques, remains largely unexplored and are the topics of this paper.…”