PSI - Issue 64

Emanuele Gandelli et al. / Procedia Structural Integrity 64 (2024) 685–692 Emanuele Gandelli / Structural Integrity Procedia 00 (2019) 000 – 000

686

2

These, usually referred to as "white noise", are purely stochastic, broadband in nature and have a negligible intensity. OMA tools are nowadays widely employed worldwide for the Structural Health Monitoring (SHM) of civil structures since capable to detect even small variations of their dynamic properties that often indicate the onset of structural damages. Among OMA techniques, "Frequency Domain Decomposition" (FDD) is a powerful tool that allows to extract structural systems’ natural frequencies and relevant mode shapes by analyzing ambient vibration time histories (Brincker and Ventura, 2015). Over the past two decades, FDD has found successful application in investigations concerning various civil structures including bridges (Cornaggia et al., 2022), historic cathedrals (Gentile et al., 2019 a), and masonry towers (Gentile et al., 2019-b). Recent studies have employed OMA for identifying flexural cracks in prestressed reinforced concrete (r.c.) beams. However, most investigations have primarily focused on assessing changes in fundamental frequency and modal shape in presence of significant damages (Zhou et al., 2010, Limongelli at al., 2016). In (Pisani et al., 2021), for instance, it is noted that changes in modal parameters for prestressed beams are only noticeable when the structure nears ultimate conditions, limiting the utility of this information for monitoring and construction management decisions. Similarly, in (Limongelli et al., 2016) OMA were employed to explore variations in fundamental frequency due to notable flexural damage, as indicated by residual beam deflection (i.e., after vertical load removal) of approximately 1.5 mm. In (Zhao et al., 2021) OMA were used to identify and locate flexural cracks on a prestressed concrete box girder and concluded that traditional modal parameters, such as natural frequencies, mode shapes, and modal damping ratios, lack sensitivity in reflecting varying levels of damage. In (Zhou et al., 2010) the same technique was exploited to detect and localize artificially induced flexural damage (spalling damage) in the girder by removing cubic portions of concrete at twelve different locations on the extrados. Unlike abovementioned past studies, this research delves into the dynamic identification of prestressed beams affected by shear cracks and/or almost negligible (i.e., fully reversible) flexural damages. Concerning the dynamics of prestressed concrete beams with bonded strands, some authors agree that their fundamental frequency is scarcely influences even by severe decreases (up to -40%) of design compression load (Deak, 1996). Otherwise, slight reductions in natural frequencies with increased compression levels ( ) are observable for beams with unbonded strands (external load). Within the same study (Deak, 1996), this formulation has been proposed by Jain and Goel: 1 =√ 1 ( 4 4 − 2 2 )⁄2 (1) being: the length of the beam; is the mass per unit length; is the elastic modulus of the concrete; is the equivalent homogeneous elastic inertia moment of the full cross-section (assumed to be fully integer); the equivalent overall (concrete plus steel rebars) area of the beam cross-section. It is wort noting that a slight variation of Eq. (1) was proposed by Dell’Asta and Dezi (Deak, 1996). More recent experimental studies do not provide a unique and clear solution of this drawback. Certain authors (Bonopera et al., 2019; Frizzarin et al., 2019) have contended that the natural frequency tends to remain relatively stable even when variations of the prestress force are introduced. Conversely, according to other studies (Miyamoto et al., 2000, Jaiswal et al., 2008), the fundamental frequency does change with prestressing force, but it is also influenced by additional parameters and/or phenomena such as tendon eccentricity and the development of micro-cracks. This renders questionable the employment of the fundamental frequency as a reliable parameter for accurate evaluations of the actual prestress level of the beam. Within this framework, the primary objectives of this study are: (1) employ, for the first time, Operational Modal Analysis (OMA) procedures for identifying the dynamic properties of Prestressed Concrete Beams (PCBs) affected by an “almost pure” shear cracks pattern (beam-1) or nearly negligible, i.e., reversible, flexural damages (beam-2); (2) dispel the doubts, through an ad-hoc set of experimental tests, regarding suitability of the fundamental frequency as a parameter for the assessment of residual tendons’ prestressing load . 2. Case-study beams This study examines two different case-study beams, named "beam-1" and "beam-2." These are full-scale beams, 10 m long and 80 cm high, with an I-section made of concrete C50/60 (i.e., compressive cubic strength of 60MPa). Fig. 1-a,b illustrates the cross-sections of the beams: Beam-1 has 6 x S15 bonded strands subjected to an initial design