Dynamic behavior of tall buildings under wind: insights from full‐scale monitoring

Abstract

SUMMARYThe wind-induced response of tall buildings is inherently sensitive to structural dynamic properties like frequency and damping ratio. The latter parameter in particular is fraught with uncertainty in the design stage and may result in a built structure whose acceleration levels exceed design predictions. This reality has motivated the need to monitor tall buildings in full-scale. This paper chronicles the authors' experiences in the analysis of full-scale dynamic response data from tall buildings around the world, including full-scale datasets from high rises in Boston, Chicago, and Seoul. In particular, this study focuses on the effects of coupling, beat phenomenon, amplitude dependence, and structural system type on dynamic properties, as well as correlating observed periods of vibration against fi nite element predictions. The fi ndings suggest the need for time-frequency analyses to identify coalescing modes and the mechanisms spurring them. The study also highlighted the effect of this phenomenon on damping values, the overestimates that can result due to amplitude dependence, as well as the comparatively larger degree of energy dissipation experienced by buildings dominated by frame action.

Wind-induced vibrations of buildings: role of transient events

“…In fact, this finding is entirely consistent with the levels of damping observed in stationary random decrement analyses of low-amplitude sway modes by Kijewski-Correa and Pirnia (2007). That study found that as amplitude levels dropped, these sway modes could show particularly large damping values.…”

“…The amplitude range canvassed here surpasses previous analyses of this building and confirms the reduction in natural frequency with amplitude initially observed by Kijewski-Correa and Pirnia (2007). Still there is appreciable scatter in these instantaneous frequency estimates, which in part may be attributed to the variance inherent to this system identification method that cannot benefit from averaging, but also to the dynamics of the problem, as the next section will demonstrate.…”

“…While wavelets have been used previously by the authors to visualise transient features (Bentz and Kijewski-Correa, 2009a;Kijewski-Correa and Pirnia, 2007), this study takes the wavelet analysis a step further by actually performing system identification within the wavelet framework. The composite building studied here, whose structural details and instrumentation are described in greater detail by Pirnia et al (2007), is instrumented with three orthogonal pairs of accelerometers attached to girders on its 64th floor.…”

“…Again this is consistent with the observations in Figure 3(b), which further suggest that the benefits of this exchange can transfer energy from torsion to sway and vice versa. It is not surprising then to note that Kijewski-Correa and Pirnia (2007) documented higher uncertainty in the estimates of damping in the y-mode, even in stationary wind events, likely to be due to the effect of coupling. …”

System Identification of Constructed Facilities: Challenges and Opportunities across Hazards

“…This form of local averaging has been shown to improve RDT performance by minimizing trigger sensitivity, while still permitting documentation of the degree of amplitude dependence in viscous damping ratios of several tall buildings and the effect of amplitude-dependent frequency on not only equivalent viscous damping, as shown in Figure 1, but also as a facilitator of significant lateral-torsional beat phenomena at modest wind speeds [Kijewski-Correa and Pirnia, 2007].…”