Wind-sensitive Structures Research Articles

Time domain calculations of buffeting response for wind-sensitive structures

Abstract A linearised wind loading model is developed in time domain for an element at an arbitrary attitude in the flow. The fluctuating loading is linked to a FEM program (USFOS), based on the so-called structural unit method. The initial steps of establishing multi-dimensional correlated time series of the flow from one point spectra and coherence functions are obtained by Monte Carlo simulations and regular FFT-procedures. Examples of a rigid beam, a cantilevered beam, a simply supported beam and a T-shaped bridge, have been studied. It has been focused on the effects of the decay parameter in a simple root-coherence description and on the size of finite elements in the computational model. An example of non-linear dynamic response has been included by adopting a bilinear moment–curvature relationship for the spring of the T-shaped model.

Damping in structures: its evaluation and treatment of uncertainty

Abstract This paper concerns the damping in structures with emphasis on treatment of inherent uncertainty in its prediction and estimation. Material or structural damping is addressed as well as damping due to the aerodynamic and hydrodynamic forces of the fluid surrounding the structure. The reported data base on damping information is examined in light of wind sensitive structures that rely heavily on damping for their performance under winds. The basic techniques for estimation of damping from response time histories are reviewed, and the random decrement technique is considered in some detail. The implications of the uncertainty of damping on system response are analyzed in terms of a perturbtion technique, second-moment analysis and Monte Carlo simulation. Several simple illustrative examples are provided throughout the text.

The importance of convective gusts

AbstractThis paper examines the practical importance of convective gusts for operators of wind‐sensitive systems and for designers of Wind‐sensitive structures. The mechanics of gusts associated with turbulence (driven by momentum) and convective eddies (driven by buoyancy) are discussed‐Statistical analysis of gusts and gust factors is presented. This shows that extreme gust factors are always produced by the convective mechanism whereas extreme gusts can be produced by either mechanism. Severe convective gusts are shown to have important consequences for operators of Wind‐sensitive systems because they defy precise prediction, they can be isolated in their occurrence, they can occur in quite moderate mean winds and they can be associated with sudden changes of wind direction. Extreme convective gusts are of less importance to designers of wind‐sensitive structures. They are shown to have a relatively small influence on the magnitude of the predicted extreme gusts taken as design criteria for such structures. Nevertheless, there is a significant chance that when such extreme gusts occur, they are produced by the convective mechanism. When this occurs, the predicted aerodynamic effect may be underestimated, to some extent, because of inertial forces which change the nature of the flow and may increase the loading.

Stochastic gust response of microwave lattice towers

Microwave lattice towers of heights up to about 100 m are constructed in large numbers for the requirement of an increasing network of electronic communication systems. Wind load, which is basically dynamic in nature, is the main design force for these free-standing lattice towers. The dynamic response of wind sensitive structures subjected to a random wind force can be computed using stochastic analysis. In this paper, a spectral approach for computation of the along-wind response and the gust response factor of microwave lattice towers is presented. A computer program has been developed for the calculation of the gust response factor, the along-wind response of lattice towers and wind characteristics using this spectral method. The gust response factor computed for a microwave lattice tower of 101 m height by this method is compared with the values calculated using the formulae recommended in the Indian, the Australian, the British and the ASCE Standards.

Recent approaches to extreme value estimation with application to wind speeds. Part I: the pickands method

The last 15 years have seen the development of a new body of theory and of a new generation of statistical estimation procedures applicable to extreme data. Essentially these approaches emphasize the primacy of the information concerning the largest of the extremes (the tails), as opposed to information inherent in the bulk of the extreme data (tile body of the distribution). One potentially attractive feature of some of these approaches is that, in principle, they make it possible to ascertain whether the distribution thai best fits a given set of extreme data has a finite tail. Whether extreme ~ind speed distributions have finite tails is a matter of great interest in structural engineering since the answer to this question may affect significantly the estimation of safety factors for wind-sensitive structures. Failure probability estimates based on the assumption that wind speeds are described by the Gumbel distribution (which has infinite upper tail) were shown to be unrealistically high (Ellingwood et a1,1980; Simiu, Shaver, and Filliben, 1981). This may be the case because the probability of occurrence of very high speeds, rather than being finite as in the Gumbel model, is in fact zero, since it is obvious on physical grounds that extratropical wind speeds must be bounded. It is natural to ask whether modern statistical methods can estimate those bounds from available sets of extreme wind speed data, and if so, what is the minimum size of the data sample that would allow the estimates to be made with acceptable confidence. Tile present work is motivated by these questions. It is part of a long<erm project aimed at evaluating various estimation methods Z a the theory of extremes. We examine two such methods here: (1) a condi t iona l mean exeeedanee ( C M E ) slope estimation method (Davisson-Smith,1990); and (2) the P iekands method (Pickands,1975). Both are based on the Generalized Pareto Distribution (GPD) model for excesses over a threshold. The GPD was shown to arise as a limiting distribution for excesses over thresholds if and only if the parent distribution tends asymptotically to one of the extreme value

The effect of tuned mass dampers and liquid dampers on cross-wind response of tall/slender structures

An investigation is made of the possible application of tuned liquid column dampers (TLCD) and tuned liquid column/mass dampers (TLCMD) in reducing the cross-wind response of wind-sensitive structures. The structure is modelled as a lumped mass multi-degree-freedom system taking into account both bending and shear. The cross-wind wake excitation is modelled as a stochastic process which is stationary in time and non-homogeneous in space. A random vibration analysis utilising transfer matrix formulation is carried out to obtain response statistics. The nonlinear damping term in the fundamental equation of the tuned liquid damper is treated by an equivalent linearization technique. Numerical examples show that tuned liquid dampers, which have significant practical advantages, can achieve the same motion reduction level as the traditional tuned mass dampers if the parameters of the liquid dampers are properly selected. However, excessive liquid motion in a tuned liquid column/mass damper may reduce the effectiveness of this damper and therefore a careful selection of the frequency tuning ratios of the damper is necessary.

Control of Along‐Wind Response of Structures by Mass and Liquid Dampers

An investigation is made of the possible application of tuned liquid column dampers and tuned liquid column/mass dampers in reducing the along-wind response of wind-sensitive structures. The structure is modeled as a lumped mass multi-degree-of-freedom system taking into account both bending and shear. The wind turbulence is modeled as a stochastic process that is stationary in time and nonhomogeneous in space. A random vibration analysis utilizing transfer matrix formulation is carried out to obtain response statistics. The nonlinear damping term in the fundamental equation of the tuned liquid damper is treated by an equivalent linearization technique. Numerical examples show that tuned liquid dampers, which have significant practical advantages, are as effective as the traditional tuned mass dampers if the parameters of the liquid dampers are properly selected. However, excessive liquid motion in a tuned liquid column/mass damper may reduce the effectiveness of this damper. It is also shown that the wind-induced force- and acceleration-type responses of the structure with a damper, which is usually tuned to the fundamental frequency of the structure, should involve more than one vibration mode as higher-mode responses may become as large or even larger than the controlled-mode response.

Reliability analysis of wind-sensitive structures

The theoretical background of practical structural reliability methods encompassing numerical integration, approximate methods and simulation techniques is reviewed in the context of wind-excited structures. Their relative merits, shortcomings and limitations are addressed from the standpoint of their accuracy and computational efficiency. Quantitative reliability estimates of a concrete chimney are made, using different analysis procedures. Multiple potential failure modes are represented by the exceeding of the ultimate moment capacity at any level of the chimney height. The general bounds on the system failure probability are expressed in terms of the failure probability of the individual modes. The narrower bounds are established based on the existing theory by taking into account not only the failure probability of the individual modes, but also the joint failure probabilities in any two modes. Based on the findings of this study, it is suggested that the advanced first-order second-moment approximation and the simulation methods, which combine the Monte Carlo technique with variance reduction techniques, may provide accurate results for practical reliability analysis of wind-excited structures.

Aerodynamic response of structures with parametric uncertainties

Uncertainties associated with the load effects and dynamic characteristics of wind-excited structures have been identified and discussed. Based on the available experimental data from laboratory and field study measurements the variability of the parameter space categorized as wind environment and meteorological data, wind-structure interaction and structural properties has been assessed. The probabilistic dynamic response of a wind-excited structyre has been expressed in terms of uncertain parameters. The influence of uncertainty in these parameters has been propagated in accordance with the functional relationships that relate them to the structural response. In this study, the propagation of uncertainty has been obtained by employing the Second-Moment and Monte Carlo simulation techniques. The dynamic response of a chimney subjected to aerodynamic loads is presented to illustrate the treatment of uncertainty in the parameter space. The mean and coefficient of variation of the peak chimney response, in terms of top deflections and associated base bending moments, exhibit close agreement between the Second-Moment and simulation techniques. A sensitivity analysis has been conducted which delineates the relative significance of uncertainty in the several parameters, related to both load effects and structural characteristics, on the overall uncertainty in the aerodynamic response of a chimney. The uncertainty associated with both response components suggests a need for further improvement in the modeling of wind-structure interaction, prediction of natural frequencies and damping, and reduction in the variability of extreme wind estimates. There are immediate applications of the procedures discussed in this paper for a variety of wind-sensitive structures.

Power Plant Wind Design Problems and Practices

The wind influences the design, construction, and operation of a coal-fueled power plant in many respects. This presentation explores several design, fabrication, erection, and serviceability problems and practices. Specifically considered are shortcomings of standards with regard to wall panel and fastener design, concerns about fatigue failures at low stress levels and analytical needs for wind-sensitive structures such as steel stacks and transmission poles, requirements for quality control of materials and fabrication (including welding and galvanizing), and the need for attention to structural details for force transfer. Examples are also drawn from design of generation buildings, coal handling and substation structures; construction of silos, cooling towers, and pond liners; and behavior of chimneys and stacks in both along-wide and across-wind directions.

Aerodynamic Coefficients and Risk‐consistent Design

A simple procedure is presented for estimating the reliability of windsensitive structures whose orientation is not specified. The procedure is based on a second moment reliability approach and makes use of: (1) Aerodynamic coefficients obtained experimentally as functions of wind direction, and (2) climatological data consisting of sets of largest annual speeds associated with winds blowing from each of the 8 (or 16) compass directions. An illustration of the procedure, based on sets of actual data, is presented. It is then shown that the tools presented in the paper can be applied to develop, from directional, aerodynamic, and climatological data, nominal aerodynamic coefficients which in effect reduce wind loads with respect to those inherent in current U.S. practice, in a manner that reflects the directional dependence of both the aerodynamic behavior of structures and the extreme wind climate, while being consistent with respect to failure risk. These tools are first developed for the general case wh...

Interaction of square prisms in two flow fields

A single square prism was tested at several angles of attack in a wind tunnel in smooth and in turbulent flow. A second square prism of equal size was placed upstream with various separation distances and orientations. Proper combination of signals from sensitive pressure transducers yielded the fluctuating aerodynamic forces and moments. The mean forces and moments were determined using conventional piezometer openings in conjunction with a liquid manometer. Some error is inevitable since only three transducers were used to a side. Comparisons with available information show reasonable correspondence with the mean and fluctuating forces. Little information appears available for moments. Although the upstream prism effectively shields the downstream prism in many cases, particularly for the mean forces and moments, there are in all cases, significant values for the coefficients of fluctuating force and moment. When the two objects are offset both vertically and horizontally, particularly large mean moments occur. Use of a turbulent flow field is to be recommended for testing. Consideration of moment measurement should be of prime importance in wind sensitive structures.