Multi-scale correlation analyses of two lateral profiles of full-scale downburst wind speeds

Lateral correlation analyses are carried out in different scales for two sets of full-scale thunderstorm downburst full wind speed time series. These two sets of data were recorded at the TTU field site, Texas, USA, on June 4 and 15, 2002, respectively. The same data sets were also used in the authors’ previous paper [L. Chen, C.W. Letchford, Proper orthogonal decomposition of two vertical profiles of full-scale nonstationary correlated downburst wind speeds, J. Wind Eng. Ind. Aerodyn., 93(3)(2005) 187-266.], which presents a methodology to model downburst vertical profiles. The statistical procedures utilized in [7], such as the POD technique, the wavelet shrinkage for time-varying mean inference, the two-stage weighted moving-average method for time-varying variance inference and the Priestley's EPSD estimator, are also employed in this paper and their details are thus omitted. While in [7] the authors employed the proper orthogonal decomposition (POD) to reduce the dimensions, POD in this article is employed to measure the degree of the correlation between two deterministic time series. Conventionally, wind speeds are decomposed to a time-varying mean speed and a nonstationary fluctuating speed by virtue of wavelet shrinkage [7]. Time-varying mean speeds are in the largest scale, which are essentially due to translations of storms and radial velocities of spreading flow. It was observed that the time-varying standard deviations [7] of fluctuating speeds are basically driven by time-varying mean speeds other than small-scale turbulence. Therefore, time-varying standard deviations are considered in the medium scale. Finally, the modulated fluctuating speeds are viewed in the smallest scale and due to small-scale turbulence. It is found that (1) the two lateral profiles are fully correlated and almost evenly spatially distributed in the largest and medium scales with more than 96% of the total energy; (2) they are uncorrelated in the smallest scale since the measurement locations are at least 263 m apart.     

Proper orthogonal decomposition of two vertical profiles of full-scale nonstationary downburst wind speeds[lzcl]

The paper proposes a novel analysis framework for nonstationary wind speeds, from which accumulated results can potentially lead to or enhance empirical nonstationary wind speed models such as the hybrid one for downbursts in (Eng. Struct. 26 (2004) 619). This framework is motivated by the proper orthogonal decomposition (POD) technique and consists of four major steps: separation, POD, approximation and property inference. In the first step, the original wind speed time histories are separated into their time-varying mean speeds and fluctuating speeds through wavelet shrinkage, which is a promising new tool for smoothing; fluctuating speeds are further expressed as the product of their time-varying standard deviations and normalized fluctuating speeds with unit variance. In the second step, the POD is applied to the time-varying means, standard deviations and normalized fluctuation. In the third step, by properly selecting the number of retained modes, the original time series can be approximated to a numerical model. Finally, wind properties such as velocity vertical profiles and turbulence vertical profiles can be calculated from the model. The POD employed herein is a specific POD for arbitrary multivariate data representation. The normalized fluctuations are characterized by both their power spectral densities (PSD) and evolutionary power spectral densities (EPSD). The concept of EPSD for nonstationary processes is briefly presented and one EPSD estimator is given. Specifically, the PSD/EPSD of an original process is related to the PSD/EPSD of the principal coordinate process obtained by the POD. This framework is applied to two sets of nonstationary full-scale thunderstorm downburst wind speed time series. Many appealing downburst properties are obtained and the benefit of the POD is demonstrated.     

Extreme wind load estimates based on the Gumbel distribution of dynamic pressures: an assessment

We present a contribution to the current debate on whether it is more appropriate to fit a Gumbel distribution to the time series of the extreme dynamic pressures (i.e. of the squares of the extreme wind speeds) than to fit an extreme value distribution to the time series of the extreme wind speeds themselves. It has been shown that the use of time series of the extreme dynamic pressures would be justified if the time series of the wind speed data taken at small intervals (e.g. 1 h) were, at least approximately, Rayleigh-distributed. We show that, according to sets of data we believe are typical, this is not the case. In addition, we show results of probability plot correlation coefficient (PPCC) analyses of 100 records of sample size 23 to 54, according to which the fit of reverse Weibull distributions to largest yearly wind speeds is considerably better than the fit of Gumbel distributions to the corresponding largest yearly dynamic pressures. We interpret the data and results presented in the paper as indicating that there is no convincing support to date for the hypothesis that the Gumbel distribution should be used as a model of extreme dynamic pressures.     

Analysis of height variations of sodar-derived wind speeds in Northern Spain

The Weibull distribution has been widely used in wind speed analysis. Ten-min averages from a sodar placed in an extensive plateau in the North of Spain were considered. The measuring period was April 2001. Daily wind speed evolution has been analysed and a sharp contrast between day and night has been obtained. Contrasting behaviour between surface and more distant levels has also been seen. The strong convection during the day and the stratification stability during the night were responsible for this well-defined pattern. Four wind speed intervals for a typical wind turbine were considered. Low wind speeds showed no directional character, although moderate winds came from two prevailing directions due to the synoptic weather systems affecting the Iberian Peninsula with a 5-6 day period. Ground influence was present only at lowest levels. As a result, the behaviour of wind speed derived from surface data must be avoided. Although moderate winds were frequent, a persistence analysis revealed the low number of lasting runs. The wind speed power law was also analysed and the usual one-seventh expression was proved not to be valid. The Weibull parameters were calculated by four methods: linear regression by cumulative frequency, moments, maximum likelihood and quartiles. The equivalence was in general clear. Finally, height analysis revealed that the shape parameter was around two whereas major differences were reached for the scale factor. A successful fit with the height was proved for the latter. Finally, a well-defined daily evolution was obtained, indicating that selection of a site for energy generation purposes must be preceded by a temporal study based on direct height measurements.     

Automated extraction and classification of thunderstorm and non-thunderstorm wind data for extreme-value analysis

Design wind loads are partly based on extreme value analyses of historical wind data, and limitations on the quantity and spatial resolution of wind data pose a significant challenge in such analyses. A promising source of recent wind speed and direction data is the automated surface observing system (ASOS), a network of about 1000 standardized US weather stations. To facilitate the use of ASOS data for structural engineering purposes, procedures and software are presented for (a) extraction of peak gust wind data and thunderstorm observations from archived ASOS reports, (b) classification of wind data as thunderstorm or non-thunderstorm to enable separate analyses, and (c) construction of data sets separated by specified minimum time intervals to ensure statistical independence. The procedures are illustrated using approximately 20-year datasets from three ASOS stations near New York City. It is shown that for these stations thunderstorm wind speeds dominate the extreme wind climate at long return periods. Also presented are estimates based on commingled data sets (i.e., sets containing, indiscriminately, both non-thunderstorm and thunderstorm wind speeds), which until now have been used almost exclusively for extreme wind speed estimates in the US. Analyses at additional stations will be needed to check whether these results are typical for locations with both thunderstorm and non-thunderstorm winds.     

Extreme value frequency analysis of wind data from Isfahan, Iran

Estimating maximum wind speed is an essential task in many fields of environmental and engineering risk analysis. This study used prevalent westerly annual maximum wind speeds for the period of 1983-1998 for East Isfahan station in Isfahan Province, Iran. The frequency analysis of AM data wind speeds obtained by averaging the wind data over some chosen averaging periods showed that extreme value Type I distribution is the best distribution for 15, 30, 60 and 120 min wind durations. The frequency and average corresponding duration were then plotted. This plot gives the average wind duration and speed for any given return period.     

Extreme wind speeds for various return periods during rainfall

Extreme wind speeds for various return periods during rainfall at six typical meteorological stations in Japan were estimated using the statistical method for extremes.The annual maximum wind speeds recorded from 1961 to 1980 at these stations were used for analysis. The annual maximum wind speeds were classified according to the amount of precipitation, and they were found to be well modelled by the extreme type-I distribution. Multiple regression analysis employing two parameters, i.e., annual maximum wind speed and hourly precipitation, was used to fit the type-I distribution.     

Wind characteristics of a strong typhoon

The wind characteristics of a strong typhoon (Typhoon Maemi 2003) are analyzed on the basis of 10 min wind speed samples. The wind speeds were measured simultaneously by nine vane and seven sonic anemometers at a height of about 15 m. Turbulence intensity and scale, gust factor, peak factor, decay factor of the coherence function, probability distribution function, power spectrum, and their variations with wind speed are obtained. Wind-direction-dependent analysis is conducted on the wind characteristics. Turbulence intensity decreases with wind speed and remain almost constant when the wind speed becomes high. The averaged values of gust factor and peak factor are 1.6 and 3.3, respectively. The spatial cross correlation and decay factor of the coherence function increase slightly with wind speed. The probability density function of fluctuating wind speed of a strong typhoon follows a Gaussian distribution, and the power spectrum of strong wind can be expressed by a Karman-type spectrum at the low frequency. The wind characteristics of this strong typhoon are shown to be very similar to those of non-typhoon winds.     

Exact and general FT1 penultimate distributions of extreme wind speeds drawn from tail-equivalent Weibull parents

We present a contribution to the current debate on methodologies for assessing extreme wind speeds. We review long-established extreme theory in the light of recent developments and show that it is unnecessary, indeed inappropriate, to resort to asymptotic models. We develop Cramér's method for the general penultimate distribution of extremes for the case of Weibull parents into a new model for extreme wind speeds that avoids the issues of asymptotic convergence and the associated errors. We demonstrate that the 30-year record of hourly mean wind speed at Boscombe Down, UK, has right-tail equivalence to the Weibull distribution and use this record to demonstrate the new model.     

Numerical simulation of extreme winds from thunderstorm downbursts

In (Chen, L., 2005. Vector time-varying autoregressive (TVAR) models and their application to downburst wind speeds. A PhD Dissertation, Texas Tech University, Lubbock, Texas.), a nonparametric deterministic–stochastic hybrid (NDESH) model was proposed for characterizing and simulating nonstationary thunderstorm downburst wind fields, and two sets of full-scale downburst wind speed records were fitted in this model. This paper aims to build an empirical numerical model for downburst wind fields according to the NDESH model and based on the results in (Chen, L., 2005; Chen, L., Letchford, C.W., 2005. Proper orthogonal decomposition of two vertical profiles of full-scale nonstationary downburst wind speeds. Journal of Wind Engineering and Industrial Aerodynamics 93(3), 187–216; Chen, L., Letchford, C.W., 2006. Multi-scale correlation analyses of two lateral profiles of full-scale downburst wind speeds, Journal of Wind Engineering and Industrial Aerodynamics 94(9), 675–696.), so that correlated downburst wind speeds can be synthesized from this empirical model.     

Evaluating and modelling the response of an individual to a sudden change in wind speed

The response of an individual to a sudden increase in wind velocity is important in terms of wind comfort and wind safety. This paper is concerned with the latter issue and outlines a series of physical and numerical experiments undertaken in order to evaluate the response of an individual to a sudden change in wind speed. The physical experiments were undertaken in the dynamic circuit of the Jules Verne Climatic Wind Tunnel at CSTB in Nantes, France and subjected 31 people to wind speeds up to 20 m/s. In all cases the wind speed increased from a mean value of zero to the target value in approximately 0.2 s. The wind speed required to cause loss of balance is shown to be a function of orientation and weight.     

Time series analysis of wind speed using VAR and the generalized impulse response technique

This research examines the interdependence in time series wind speed data measured in the same location at four different heights. A multiple-equation system known as a vector autoregression is proposed for characterizing the time series dynamics of wind. Additionally, the recently developed method of generalized impulse response analysis provides insight into the cross-effects of the wind series and their responses to shocks. Findings are based on analysis of contemporaneous wind speed time histories taken at 13, 33, 70 and 160 ft above ground level with a sampling rate of 10 Hz. The results indicate that wind speeds measured at 70 ft was the most variable. Further, the turbulence persisted longer at the 70-ft measurement than at the other heights. The greatest interdependence is observed at 13 ft. Gusts at 160 ft led to the greatest persistence to an “own” shock and led to greatest persistence in the responses of the other wind series.     

Facade flame height ejected from an opening of fire compartment under external wind

This paper presents an experimental study and analysis on the facade flame height ejected from an opening of fire compartment under external wind. Experiments are carried out in a reduced-scale model consisting of a cubic fire compartment with a vertical facade wall. An opening is designed at the center of one sidewall of the fire compartment at the facade side and subjected to external wind (normal to the opening) provided by a wind tunnel. The facade flame heights are measured by a CCD camera for five different openings at various fuel supply heat release rates and wind speeds. It is found that the facade flame height decreases with increasing external wind speed. A scaling analysis is performed to interpret this behavior based on the change of air entrainment into the flame from both the facing-facade and parallel-facade directions caused by the external wind flow. A global model incorporating the external wind speed, the two characteristic length scales of the opening as well as the dimensionless excess heat release rate is developed for describing the facade flame height of various conditions. The proposed model correlates the experimental data well.     

Determination of topographical exposure factors in complicated hilly terrain

The approach to calculating design wind speeds outlined in a recent publication of the Engineering Science Data Unit makes use of a topographical exposure factor to describe variations of wind speed over terrains other than flat, level terrains. This communication presents an analysis of several sets of experimental results from which values of this factor have been derived for rather complicated hilly terrain. These results are taken from met. office data, from the results of wind measurements made by British Rail in Cumbria, and from topographical model tests of the area surrounding the British Rail measurement sites, carried out at Oxford University. These measurements show that this factor covers only a rather narrow range of values between 0.8 and 1.2 for sites ranging from sheltered valley bottom sites to very exposed hilltop sites.     

The methodology for aerodynamic study on a small domestic wind turbine with scoop

The aim of this study is to investigate the possibility of improving wind energy capture, under low wind speed conditions, in a built-up area, and the design of a small wind generator for domestic use in such areas. This paper reports the first part of this study: the development of the methodology using physical tests conducted in a boundary layer wind tunnel and computer modelling using commercial computational fluid dynamics (CFD) code. The activities reported in this paper are optimisation of a scoop design and validation of the CFD model. The final design of scoop boosts the airflow speed by a factor of 1.5 times equivalent to an increase in power output of 2.2 times with the same swept area. Wind tunnel tests show that the scoop increases the output power of the wind turbine. The results also indicate that, by using a scoop, energy capture can be improved at lower wind speeds. The experimentally determined power curves of the wind generator located in the scoop are in good agreement with those predicted by the CFD model. This suggests that first the developed computer model was robust and could be used later for design purposes. Second the methodology developed here could be validated in a future study for a new rotor blade system to function well within the scoop. The power generation of such a new wind turbine is expected to be increased, particularly at locations where average wind speed is lower and more turbulent. The further study will be reported elsewhere.     

Wind tunnel tests on the dispersion of vehicular pollutants in an urban area

Results are presented from an extensive series of wind tunnel tests to investigate the dispersion of a tracer gas emitted from a ground level source on a 1 : 125th scale model of an urban area typical of those found in the UK. Concentrations of a tracer gas, propane in air, were measured at a number of pedestrian level locations around the model using fast-response flame ionisation detectors. These were supplemented by pedestrian level wind speed measurements using hot-film anemometry. Measurements were made for changes in the wind azimuth angle and for changes to the building geometry. General conclusions were drawn about the relationships between building geometry and the pedestrian level atmospheric environment and a number of simple empirical relationships derived between the changes in the measured wind characteristics and the characteristics of tracer concentration. Both mean and fluctuating wind speeds and concentrations were measured in the experiments and subsequently quantified using a novel empirical method that may potentially be of use to both urban planners and wind tunnel engineers.     

A method to assess peak storm wind speeds using detailed damage surveys

A detailed damage survey of a single, wood-framed, structure, which had a complete roof failure during the passage of a gust front in southern Ontario, was performed. Radar data was used to estimate upper level wind speeds associated with the gust front. Details pertaining to the structural failure, including the debris field, were obtained. Wind tunnel pressure time histories, in a simulated atmospheric boundary layer, were used to establish the roof height, gust wind speed at failure. This speed was smaller than the upper level speed found from the analysis of the radar. The flight of the roof was also examined, and confirmed the wind speeds obtained from the structural analysis of the failure. The study illustrates that detailed damage surveys, which incorporate the use of wind tunnel test data and debris flight in the analysis, can shed considerable light on the details of the wind speeds at failure, reducing the uncertainty caused by the many assumptions in such analyses.     

Durations distribution of Rayleigh process with application to wind turbines

In the present paper, approximations to certain functions of the Rayleigh process distribution, beyond extreme thresholds, is derived. A close agreement between the results obtained and the corresponding results of the exact distributions confirms the validity of the present work. The derived approximations are implemented to analyze a real wind speed data and the results obtained are provided in this article. The behavior of a wind turbine (WT) is characterized by its power curve, which includes two extreme points: cut-in wind speed and cut-out wind speed. Therefore, accurate prediction of wind speed durations beyond these two thresholds will enable engineers to acquire several probabilistic statements such as the availability of electric power generation. These findings will be beneficial to design an efficient and reliable wind energy conversion system.     

Mean wind load induced incompatibility in nonlinear aeroelastic simulations of bridge spans

Mean wind response induced incompatibility and nonlinearity in bridge aerodynamics is discussed, where the mean wind and aeroelastic loads are applied simultaneously in time domain. A kind of incompatibility is found during the simultaneous simulation of the mean wind and aeroelastic loads, which leads to incorrect mean wind structural responses. It is found that the mathematic expectations (or limiting characteristics) of the aeroelastic models are fundamental to this kind of incompatibility. In this paper, two aeroelastic models are presented and discussed, one of indicial-function-denoted (IF-denoted) and another of rational-function-denoted (RF-denoted). It is shown that, in cases of low wind speeds, the IF-denoted model reflects correctly the mean wind load properties, and results in correct mean structural responses; in contrast, the RF-denoted model leads to incorrect mean responses due to its nonphysical mean properties. At very high wind speeds, however, even the IF-denoted model can lead to significant deviation from the correct response due to steady aerodynamic nonlinearity. To solve the incompatibility at high wind speeds, a methodology of subtraction of pseudo-steady effects from the aeroelastic model is put forward in this work. Finally, with the method presented, aeroelastic nonlinearity resulted from the mean wind response is investigated at both moderate and high wind speeds.