While high-rise construction serves as one of the most challenging projects undertaken by society each year, tall buildings are one of the few constructed facilities whose design relies solely upon analytical and scaled models, which, though based upon fundamental mechanics and years of research and experience, has yet to be systematically validated in full-scale. As high-rise buildings gain more prominence worldwide, their impacts upon the global society and economy will become more pronounced, necessitating a new frontier in tall building design fully equipped to address the emerging issues of performance, economy and efficiency.
As tall-building projects push the envelope to greater heights, designers are faced with the task of not only choosing a structural system to carry the lateral loads, but also covering a design that meets serviceability and occupant comfort requirements under complex wind environments. An additional limitation in tall building design is the inability to provide accurate estimates of structural damping in the design phase, which is critical to insure that the structure can meet both serviceability and habitability requirements. Although the building stiffness may be accurately quantified, inherent damping values are typically assumed in the design stage, resulting in estimates of response characteristics that may have significant inaccuracies. Thus, the accurate prediction of inherent damping for a given design becomes yet another critical consideration. (Kijewski T. and Kareem A., 2001)
A consequence of turbulence is that dynamic loading on a structure depends on the size of the eddies. Large eddies, whose dimensions are comparable with the structure, give rise to well correlated pressures as they envelop the structure. On the other hand, small eddies result in pressures on various parts of a structure that become practically uncorrelated with distance of separation. Eddies generated around a typical structure are shown in Fig.1 (P. Mendis, T. Ngo, N. Haritos et al., 2007)
In the case of high-rise construction there have been limited full-scale monitoring projects, typically undertaken following some performance concerns. Although there have been significant studies, the circumstances surrounding many of these studies prohibited the academic communityâ€™s access to the measured data. Perhaps this explains the reservations that previously prohibited extensive full-scale monitoring of buildings in the United States. By embarking on a study of buildings with proven performance, the misconceptions regarding full-scale monitoring can be arrested, reassuring owners and occupants that the presence of monitoring devices in a structure is not necessarily indicative of a troubled building, but rather is representative of a commitment on the parts of owners and the engineering community to improve the understanding of structures and thereby techniques for their design, thus improving the habitability of the built environment. In particular, the use of a simplified user interface and Internet technologies in this study enhances this initiative by permitting a select group of researchers and owners to access daily reports of building performance, actively involving them in the health monitoring of these structures. This will promote the use of health monitoring of major buildings and pave the way for the potential applications of smart materials and advanced technologies to control building performance in the United States. It is only through such a commitment to full-scale monitoring and validation that the standards for high-rise construction can advance, resulting in more efficient, reliable designs. (Hansen R, Vanmarcke E., 1979)
Key words: Tall building, Wind-induced motion, Full scale monitoring, eddies, dynamic analysis, wind tunnel testing
2.0 Research context
The primary objective of this study is to correlate the in-situ measured response characteristics of tall buildings, representing a host of typical structural systems under a wide range of wind environments, with computer-based analytical and wind tunnel predictive models for the advancement of the current state-of-the-art in tall building design. The project team is collaborating with an advisory board of leading designers, researchers and organizations, who have been involved in a wide range of projects associated with tall buildings around the world, to provide regular reviews and consultations. The end result will be the first systematic validation of existing design practice for tall buildings in the US, followed by appropriate calibrations of existing wind tunnel and analytical models, with modifications to current design practice, where applicable.
The four buildings selected for this study represent a variety of typical structural systems employed in the construction of high-rise structures, such as the steel moment-connected bundled tubular system, exemplified by Building 1. The structure is intended to behave as a vertical cantilever fixed at the base to resist wind loads, with a skeleton comprised of a structural steel frame, pre-assembled in sections and bolted in place on site. The two interior frames of the structure connect opposing faÃ§ade frames at two intermediate points reducing the shear lag effect in the flange frames and the material premium for height considerably. Foundations are comprised of straight shaft reinforced concrete caissons extending to bedrock.
The braced steel tube concept of Building 2 has proven highly efficient in high-rise construction, relying on a steel tube comprised of the exterior columns and spandrel beams as its primary lateral load-resisting system, stiffened by its diagonal members. The diagonals tie to the structural floors and corner columns, functioning as a continuous system for the transmission of axial loads. The intersection of these diagonals at the corner columns permits the direct transfer of wind shear, carried as axial loads in the web side diagonals, to the flange side diagonals. The lateral load of the structure is resisted primarily by cantilever action (80%) with frame action carrying the remainder of the load. This behavior is primarily a result of the diagonals insuring a near uniform distribution of load on the columns across the flange face, with very little shear lag. The structure rests on foundations of straight shaft reinforced concrete caissons to bedrock. (Kijewski T. and Kareem A., 2001)
While structural steel is commonly used for high-rise construction, comparative performance of reinforced concrete structures will also prove beneficial in this study. Thus, Building 3 is examined as an example of a concrete shear wall/outrigger system. The structure is rectangular at its lower levels, approaching a square shape at the 59th floor as the result of a series of plan setbacks on the north and south faces. Above the 59th floor, the building begins to taper into a sculptured cone, topped by an 82 ft. architectural spire. Four shear walls located at the core of the building provide lateral load resistance. At the 40th and 59th floors, the core is tied to the perimeter columns at two locations via reinforced concrete outrigger walls to control the wind drift and reduce overturning moment in the core shear walls. The structureâ€™s foundation utilizes reinforced concrete straight-shaft caissons extending to rock.
The next phase of the effort will be concerned with determining the responses of each of these buildings under wind events with known wind speeds and directions through detailed wind tunnel studies. While wind tunnel tests have been conducted for most of the buildings, modifications to the near field effects due to blockage or interference of neighboring buildings have changed the conditions modeled in previous wind tunnel studies. The inclusion of these changes in the surrounding environment, along with recent advances in wind tunnel testing techniques, will produce superior wind tunnel results, providing state-of-the-art wind loading and structural analyses to accompany the full-scale monitoring of the four buildings. The implementation of high-frequency force balance model tests will provide a consistent baseline for each of the buildings. These predictions will be obtained using best estimates of in-situ dynamic properties, as well as the dynamic properties obtained by analytical methods. This will provide a valuable indication of the variation of wind-induced behavior of tall buildings due to inherent uncertainties in the actual dynamic properties. These predictions will be updated as the in-situ dynamic properties of each building are determined. Wind tunnel tests will be conducted at BLWTL, since original tests in the design phase were conducted there. (Kijewski T. and Kareem A., 2001)
5.2 Dynamic analysis
The next phase of the project will focus on determining the actual dynamic characteristics of the buildings, i.e. their natural periods and modal damping. While system identification technologies have made considerable advancements, the most accurate means to estimate stiffness and damping properties require the measurement of the input to the system, in addition to response quantities at a number of locations, making such approaches ill-suited to the wind-excited problem, for which exact system inputs are not known. While traditional analysis techniques such as Fourier analysis can be invoked, this study will consider and develop alternative approaches.
The Random Decrement Technique has emerged as a popular choice for time-domain analysis of ambient wind response data, due to its ability to produce reasonable estimates of the decay signature associated with the autocorrelation function of mechanical oscillators. The decrement signatures are obtained by overlaying segments of the structural response to ambient loads that initiate with the same initial conditions, averaging out the random components of the response. The method is particularly attractive since the user defined initial condition thresholds permit referencing of dynamic properties over varying response amplitudes. (Kijewski T. and Kareem A., 2001)
Assuming amplitude-dependent levels of frequency and damping, accurate techniques for tracking the levels of damping and natural frequency can be achieved through the use of wavelet-based extraction schemes on the resulting decrement signatures. The reliability of the estimated damping and natural frequency determined by this approach may be accessed using a Bootstrap-based resampling scheme15, 16. Unlike traditional Fourier-Based Analysis, which relies on infinite basis functions of sines and cosines, the wavelet transform decomposes signals with respect to finite, localized bases, as illustrated by the example of the Morlet Wavelet in Fig. 5. Though strikingly similar to the Fourier Basis, the Morlet Wavelet is enveloped by a Gaussian Function, yielding a localized basis capable of capturing time-varying characteristics of the system. As the wavelet provides a representation optimized in both time and frequency (or scale), the analysis of multi-mode data is simplified, as the response of a given mode is contained within a distinct frequency band associated with a particular frequency over the entire time axis. Using this transform, in conjunction with the standard techniques for modal parameter estimation based on the approximation of the signal by a complex analytic signal, extraction of the time-varying phase and envelope is possible. These may be directly related to the instantaneous frequency and damping of the system analogous to the Hilbert Transform-based approach. (Kijewski T. and Kareem A., 2001)
This report will address the deficiencies in current tall building design practice by correlating the actual performance of constructed buildings with predictions made during their design, thereby providing an important missing link between predictions and actual behavior. This will be the focus of this project. This study will provide substantive information with respect to this correlation and create an important database for the development of future prediction techniques. Furthermore, it will contribute to existing international databases by providing valuable information on the dynamic characteristics of high-rise buildings, as well as revealing modifications to dynamic properties as the result of aging. The current phase of this project is focused on the instrumentation of the four high-rise buildings, which includes a GPS, accelerometers and anemometer. Following this phase, detailed dynamic analyses, wind tunnel testing and sensitivity studies will provide valuable lessons to shape the future state of the art of high-rise design. With improved understanding of tall building performance, the next generation in design can truly evolve from the perspectives of performance, economy and efficiency to advance the burgeoning development of tall buildings around the globe.