Search

Summary: The wind direction in the atmospheric boundary layer (ABL) twists with height due to the Coriolis force; this phenomenon is called the Ekman spiral. However, this phenomenon is generally not considered in the present wind load estimation of super high‐rise buildings, which may lead to an incorrect estimation and affect the safety of structures. Therefore, this study considers and analyzes the influence of the Ekman spiral phenomenon in the wind direction reduction effect (WDRE) of the wind load of super high‐rise buildings. First, this paper proposes an empirical fitting equation for the twisted wind direction angle for a height of 100–800 m according to the classical Ekman spiral theory model (CE model). Subsequently, on the basis of twisted wind, this paper proposes a method for the correction of the wind direction reduction factors (WDRFs) of strong winds considering the influence of the Ekman spiral phenomenon in the design wind load estimation of super high‐rise buildings with heights of 400–800 m. A high‐frequency balance force measurement test of a square‐section super high‐rise building model was performed to analyze the influence of the Ekman spiral phenomenon on the WDRE of the aerodynamic force and wind‐induced response. Three Chinese cities (i.e., Beijing, Wuhan, and Kunming) are selected as case studies to illustrate the importance and necessity of the correction method. The results demonstrate that the proposed empirical fitting equation accurately determines the twisted wind direction angle at different latitudes and altitudes. Furthermore, estimating the design wind load while considering the WDRE and neglecting the influence of the Ekman spiral phenomenon may lead to a significant underestimation of the wind load of super high‐rise buildings, rendering the designed building structure more dangerous. [ABSTRACT FROM AUTHOR]

This study presents the results of a comparative study of the wind characteristics of three typhoons presenting similar characteristics, based on stationary and nonstationary models. The original data were collected at four different heights along the seaside of the Fujian Province (China), where the typhoons passed through. First, the run-test method and discrete wavelet transform were employed to evaluate the stationarity and extract the time-varying mean wind speed after data filtering. Then, the gust factor, turbulence intensity and turbulence integral scale were compared. The results demonstrated that the wind characteristics described by the nonstationary model were more centralized and stable than those obtained by the stationary model. Lastly, the power spectral density and evolutionary power spectral density (EPSD) were compared, revealing that the von Karman spectra fitted well the measured spectra. In addition, two methods for the analysis of nonstationary wind spectra were compared. The direct extension from stationary model to nonstationary model was found not to be reasonable from the results of this comparison. Considering the instantaneous energy concentration shown in the EPSD, a nonstationary approach is recommended when analyzing near surface typhoon wind data. [ABSTRACT FROM AUTHOR]

This paper proposes a novel method for estimating the evolutionary power spectral density (EPSD) of a nonstationary process based on a single sample. In the proposed method, a sample of a nonstationary process is decomposed into several components with a new binomial fitting decomposition (BFD). The EPSD of each component can be estimated using a newly proposed time-varying standard deviation estimation method and short-time Thomson multiple-window spectrum estimation method. The EPSD of the analyzed nonstationary sample is obtained by combining the EPSDs of all components. Via a comprehensive numerical study, the applicability of the proposed EPSD estimation method (for estimating the EPSD of a nonstationary process) is analyzed and compared with those by the Priestley method and wavelet-based method. The numerical results indicate that the estimated EPSD by the proposed method is more consistent with the corresponding theoretical one than those by the other two methods. Finally, the EPSDs of Storm Ampil, measured atop the Shanghai World Financial Center, are analyzed by the proposed method. [ABSTRACT FROM AUTHOR]

Stochastic wind velocity fields are usually simulated as N-variate stationary Gaussian processes by using the spectral representation method (SRM). However, the temporal statistics estimated from one SRM-simulated sample wind process cannot coincide with the target characteristics; the disagreements can be described by the bias and stochastic errors. For controlling the errors efficiently, this paper assesses the errors produced by both the Cholesky decomposition-based SRM (CSRM) and the eigendecomposition-based SRM (ESRM). The SRM is revisited first, followed by computing the temporal mean value, correlation function, power spectral density (PSD), and standard deviation of the SRM-simulated wind process. It is shown that the temporal correlation function and standard deviation are Gaussian, while the temporal PSD is non-Gaussian. Further, as mathematical expectations and standard deviations of the corresponding temporal estimations, the bias errors and stochastic errors of all the first- and second-order statistics are obtained in closed form for both the CSRM and the ESRM; the closed-form solutions are then verified in the numerical example. More importantly, this example is employed for taking a clear look at and making a comparison between the stochastic errors produced by the CSRM and by the ESRM; observations suggest that (1) in sum, the ESRM produces smaller stochastic errors than the CSRM and (2) if the ESRM is employed, stochastic errors will be distributed to each component of the wind process in a more uniform pattern. Moreover, some practical approaches are proposed to control the stochastic errors. [ABSTRACT FROM AUTHOR]

This paper presents coherency matrix-based proper orthogonal decomposition (CPOD), which boasts an improved form of power spectral matrix-based POD. Original formulas of POD are derived for a multi-correlated stationary stochastic vector process, involving the eigen-decomposition of its cross power spectral density matrixes. These matrixes are substituted by the product of the coherency matrixes and the diagonal auto spectral matrixes, resulting the CPOD on the basis of the eigen-decomposition of only coherency matrixes. The continuous formulas are immediately discretized into the CPOD-based spectral representation method that serves as a simulation tool for stochastic loading. Application of this method to a stochastic wind velocity field is included as a numerical example. The CPOD eigenvalues and eigenvectors (modes) of this wind field are computed and described in detail in comparison to the corresponding items of POD, which indicates a more straightforward link between CPOD modes and distribution patterns of the wind field. [ABSTRACT FROM AUTHOR]

Abstract: Effects of multi-mode coupling should be taken into account in computing the resonant buffeting response of some kinds of flexible structures with low damping and concentrated modes. In this paper, a new concept of “mode coupling factor” for computation of coupling effects between multi-mode resonant responses of the structures is proposed. On the base of the mode coupling factor, a modified SRSS method for computation of the resonant response contributed by multi-modes and their coupling effects of the structures is further raised. The roof structure of Shanghai Southern Railway Station is then taken as the case study to indicate the application and to verify the precision of the mode coupling factor and the modified SRSS method. The computaion results indicate that the mode coupling factor can quantitatively describe the contribution of mode coupling to the resonant response; and the modified SRSS method can make the computation of structural resonant response with consideration of mode coupling effects simpler. [Copyright &y& Elsevier]

A novel method for ergodic wind field simulation is proposed by incorporating proper orthogonal decomposition-based spectral representation method and double-indexing frequencies. The spectral representation method based on proper orthogonal decomposition (POD) is described in detail firstly. By introducing the double-indexing frequencies, a new simulation algorithm is proposed to generate not only stationary but also ergodic wind fields. The first and second statistical moments of the ensemble from the simulated fields are proved to be equal to the corresponding target ones. Also, it is shown that the simulation algorithm can generate ergodic fields, that is, the temporal means and correlation functions of the simulated fields and the target fields are equal. The algorithm is computationally very efficient due to the fast Fourier transform (FFT) technique. Some extended discussion focused on the accuracy and efficiency of the algorithm is included, in comparison to the original spectral representation method. A numerical example involving simulation of turbulent wind velocity fluctuations on the Longtan river bridge is presented, by which capabilities of the proposed method are demonstrated. [ABSTRACT FROM AUTHOR]

Flutter derivatives show great importance on analysis of bridge wind-induced vibration. They can be extracted through wind tunnel tests and CFD simulation. In this paper, a novel approach based on the CFD discrete time aerodynamic model is presented to identify flutter derivatives. This method employs the 3211 multi-step as input of velocity to excite CFD model, which can provide aerodynamic forces acting on the body through CFD simulation. An aerodynamic model in discrete time domain, defined on frequency range of interesting, can be developed based on the input-output relationship. Then the obtained aerodynamic model can be used to simulate aerodynamic response for any harmonic input with suitable frequency. Finally, flutter derivatives can be identified based on input and output time histories. The present method can extract flutter derivatives under arbitrary reduced wind speed, requiring respectively only once round of CFD Simulation for heaving or pitching motion. Identification of flutter derivatives of a thin plate is carried Out by using the present method, reasonable agreements between the identified flutter derivatives and those from wind tunnel tests validate the efficiency of the present approach. [ABSTRACT FROM AUTHOR]

Flutter derivatives and aerodynamic admittances provide basis of predicting the critical wind speed in flutter and buffeting analysis of long-span cable-supported bridges. In this paper, one popular stochastic system identification technique, covariance-driven stochastic subspace identification (SSI in short), is first presented for estimation of the flutter derivatives and aerodynamic admittances of bridge decks from their random responses in turbulent flow. Numerical simulations of an ideal thin plate are adopted to extract these aerodynamic parameters to evaluate the applicability of the present method. Then wind tunnel tests of a streamlined thin plate model and a Π type blunt bridge section model were conducted in turbulent flow and the flutter derivatives and aerodynamic admittances are determined by the SSI technique. The identified aerodynamic parameters are compared with the theoretical ones and the results indicate the applicability of the current method. [Copyright &y& Elsevier]

This paper proposes a new method for estimating the peak factor and gust factor of nonstationary wind. In this method, the probability density function (PDF) of the largest value of a nonstationary wind speed is fitted using the PDF of the largest value of an equivalent stationary wind speed; subsequently the mean wind speed and standard deviation of the equivalent stationary wind speed, which are respectively defined as equivalent mean wind speed and equivalent standard deviation, are used to normalize the largest value of the nonstationary wind speed to calculate its gust factor and peak factor. The proposed method can be utilized in the field measurement of nonstationary wind speed to obtain peak factor and gust factor which might be applicable for predicting the largest value of a new nonstationary wind speed only using its time-varying mean wind speed and time-varying standard deviation. The applicability of the proposed method is verified using numerical simulation and wind speed data of three typhoons, which were measured atop the Shanghai World Financial Center (SWFC) at a height of 497 m. The three typhoons' peak factor and gust factor estimated by the proposed method are also analyzed in this study. • A new method for estimating the peak factor and gust factor of nonstationary wind. • Peak factor and gust factor by the new method are applicable for predicting the largest value of a new wind speed. • Verification of the applicability of the proposed method using numerical simulation and measured wind speed data. • Estimation of the peak factor and gust factor of three observed typhoons using the proposed method. [ABSTRACT FROM AUTHOR]

Summary: Lattice towers are among the most vulnerable structures during typhoons. In this study, the wind engineering research field base near Shanghai Pudong International Airport was used to measure the mean wind speeds and directions at a height of 10 m, as well as the accelerations atop a 40‐m high lattice tower during Typhoons Neoguri and Nakri to investigate the characteristics of the near‐ground wind and the responses of the lattice tower. Moreover, an aeroelastic model of the lattice tower was generated by the discrete stiffness method. Using the wind tunnel test of the aeroelastic model, the root mean square accelerations of the tower were studied with different wind speeds and directions. The responses of the tower in the across‐wind direction were found always higher than those in the along‐wind direction, which meant the vortex shedding of each single cylinder had an impact on the responses of the tower in the across‐wind direction. The test results were compared with those of field measurements to evaluate the accuracy of the aeroelastic model, and the two were generally in a good agreement. Thus, the aeroelastic model could precisely simulate the dynamic characteristics of a prototypical lattice tower using the discrete stiffness method. [ABSTRACT FROM AUTHOR]

Modelling the complete probabilistic information of wind speed processes, which can describe the joint probability distributions of wind speeds in the time domain, is significant and challenging. In this study, a joint probability distribution model of fluctuating wind speeds of monsoons is directly estimated from field-measured wind speed samples. The fluctuating wind speeds are modelled by a harmonisable process in the frequency domain. The marginal probability distribution of the wind speed frequency components is represented by a derived mixture distribution model consisting of the Gaussian distribution and generalised Pareto distribution. The probabilistic dependence among the frequency components is modelled by the D-vine copula distribution. The results show that the measured samples are nonstationary in the time domain and non-Gaussian in both time and frequency domains. The probabilistic dependence among the frequency components is moderate but cannot be ignored. The estimated joint probability distribution model can characterise the mean value, standard deviation, kurtosis coefficient, probability density functions (PDFs) of the measured samples and the PDF of the measured maximum wind speed in the time domain. The wind skewness coefficient and the PDF of the minimum wind speed from the obtained joint probability distribution model have certain differences from the measured results. • A joint probability distribution model of monsoon fluctuating wind speeds is estimated from field-measured wind speeds. • The measured fluctuating wind speeds are nonstationary and non-Gaussian. • The probabilistic dependence among the wind speed frequency components is moderate but cannot be ignored. [ABSTRACT FROM AUTHOR]

In this paper, wind tunnel tests and CFD (Computational Fluid Dynamics) simulations are conducted to measure and simulate the near-roof wind speed and friction velocity on gable roofs. By comparing the results of Irwin sensor wind tunnel tests with CFD simulations of different turbulence models, it is found that the Realizable k - ε turbulence model performs better. The CFD simulations and wind tunnel tests show that the mean near-roof wind speed and the mean friction velocity on the windward and leeward gable roofs have the same variation trend with the slope. Moreover, the mean near-roof wind speed and the mean friction velocity reach the maximum on the windward roof when the slope is 20°, the mean wind speed and the mean friction velocity on the leeward roof decrease with the increase of the slope. Finally, based on the snow transport rate on the windward roof with different slopes, it can be obtained that the snow transport rate on the windward gable roof with 20° is the largest among the four slopes selected in this paper. Meanwhile, when the roof slope is 20°, it is most likely to lead to the unbalanced snow load on the windward side and leeward side. • The wind speed and friction velocity on roof surface are obtained by wind tunnel test and CFD simulation. • The influence of slope on roof wind speed and friction velocity is analyzed. • The effect of slope on snow transport rate and unbalanced snow load on the roof is studied. [ABSTRACT FROM AUTHOR]

Accurate estimation of wind coherence based on field-measured is a fundamental issue in which sufficient frequency resolution at low-frequency range and low variance at high-frequency range should be satisfied together. With essential limitations, commonly-used Welch and multi-taper (MT) methods may not meet this requirement. A new multi-taper S-transform (MTST), which utilizes specially designed orthogonal time-frequency windows, has a great potential for wind coherence estimation. In this study, a comprehensive comparison of the applicability and accuracy of different coherence estimation methods is firstly presented. Based on simulated processes, the estimation coherences, including co-coherence and quad-coherence, are compared with different window widths and taper numbers in Welch and MT, respectively. Constrained by the parameter setting, existing commonly-used methods cannot achieve the trade-off between variance reduction and frequency resolution improvement. MTST has good performance for single-sample coherence estimation over the entire frequency domain. Subsequently, detailed model suitability evaluations of five wind co-coherence models are also performed. Three similar typhoons captured simultaneously at four heights are adopted. The statistical properties of decay parameters are presented, including the effect of separation distances, wind velocity, and different estimation algorithms. These assessment results are valuable for accurate and reasonable wind resistance design. • The failure of conventional methods to guarantee both frequency resolution and accuracy coherence estimation is investigated. • A comparison against existing coherence estimation methods was conducted, including co-coherence and quad-coherence. • MTST has good performance for coherence estimation, especially for single-sample, with a variable data-frequency window. • Existing coherence models are evaluated by the estimation results of MTST and conventional method with measured typhoon data. • Hansen's theoretical model performs well with investigated data. The fitting results of decay parameter are also provided. [ABSTRACT FROM AUTHOR]

A numerical model based on Lagrangian description is developed in this study to predict wind-induced snow drifting on flat roofs, which fully considers the key physical processes of snow particle saltation. A scaled experiment of snow redistributions on the model roof was carried out in the wind tunnel for the verification and evaluation of different numerical methods selected from related literature. The optimal model is determined by comparing the predicted transport rate on the roof with estimations from empirical formulas and by comparing simulated snow distributions with experimental data. Characteristics of the saltation layer above the flat roof surface are then presented in detail based on the developed model. The rebound particles account for approximately 60% of the total transported particles above the roof, indicating the dominant role of rebound in the lift-off process for saltating snow particles. The developing snow transport along the roof span leads to the increase of surface erosion flux caused by the splash entrainment with downwind distance. The splash entrainment could thus contribute considerably to unsaturated snow transport on the flat roof, which is also found to be influenced by the approaching wind velocity and threshold friction velocity to varying degrees. • A numerical model is developed using the Lagrangian approach to investigate the mechanisms of snow drifting on flat roofs. • A scaled experiment of snow redistributions on flat roof is conducted in the wind tunnel to evaluate the simulated results. • Characteristics of the saltation layer above the flat roof are presented based on the developed model. • The role of splash entrainment and its contribution to unsaturated snow transport on flat roofs are preliminarily explored. [ABSTRACT FROM AUTHOR]

The wind characteristics of typhoon 'Muifa' such as power spectrum, cross power spectrum and coherence function are analyzed on the basis of long time measured data which is recorded respectively at the height of 10m, 20m and 40m above the ground in Pudong Shanghai, and two kinds of new cross power spectrum models are proposed based on Kaimal model. It is shown that longitudinal fluctuating wind power spectrum roughly fit with Von Karman spectrum, but there has deviation in high frequency band between lateral and vertical fluctuating wind power spectrum and Von Karman spectrum. By comparison, it can be found that measured fluctuating wind cross power spectrum is in agreement with modified Kaimal model. Averaged decay factor of 10min wind speed samples in longitudinal, lateral and vertical directions are 10.71, 7.43 and 8.63, respectively. In addition, the measured spatial coherence is consistent with Davenport's empirical formula. [ABSTRACT FROM AUTHOR]

The implementation of the probability analysis method on extreme wind loads considering the directionality and uncertainties of the wind load effect coefficients and wind speeds can yield accurate design wind loads. However, it is difficult to popularize this method among structural designers because its theory and calculation process are complicated. The directional design wind speeds for a given city, which are more suitable in the commonly used simplified method for estimating the wind loads (i.e., the sector-by-sector method), are obtained in this study via multi-parameter optimization. A multiple population genetic algorithm program is established for optimization based on the time histories of the wind pressure coefficients obtained from numerous wind pressure taps of different rigid models. For different structural aerodynamic characteristics, the proposed directional design wind speeds can be combined with the sector-by-sector method to obtain results that are closest to the overall reference values of the design wind pressures of all pressure taps under a certain safety guarantee. The applicability, economy, and safety of the sector-by-sector method with the proposed directional design wind speeds are discussed in comparison to other simplified estimation methods using examples of Des Moines, IA, USA, and Shanghai, China, for non-typhoon and typhoon-prone climates, respectively. • An MPGA (multiple population genetic algorithm) program was proposed to determine the directional design wind speeds. • The results of the MPGA program can significantly improve the effect of the sector-by-sector method. • The use of the results of the MPGA program does not affect the concision of the sector-by-sector method. • The results of the MPGA program can be easily reused to estimate the extreme wind pressures on different buildings. [ABSTRACT FROM AUTHOR]

This paper introduces a novel snow transport model for the analytical simulation of wind-induced snow transport on flat roofs. In the novel model, the differences between precipitation and deposited snow particles on roofs, including the differences in threshold friction velocity and fetch distance for saturated drifting state, are considered. Thus, snow drifting on a roof is divided into drifting without and during snowfall. The effects of terrain category and roof height are also considered using the friction velocity on roofs to compute snow transport rate instead of directly using the wind speed data reported by weather stations. Parametric analysis and case study are conducted to explain the novel model in detail and compare the results with those obtained from the model developed by O'Rourke et al. (2005). The estimated fetch distance for saturated snow drifting by the proposed approach in this study ranges from zero to thousands of meters, which is consistent with field measurement results and more reasonable than the use of constant values. Results of the case study indicate that the proposed snow transport model might be more reasonable because the effects of more factors are considered. • A novel snow transport model considering the effects of wind exposure on snow load is introduced. • Snow drifting is divided into drifting without snowfall and drifting during snowfall. • Methods for estimating fetch distance for saturated drifting are developed. • Impacts of terrain and roof height are taken into consideration. [ABSTRACT FROM AUTHOR]

The insight into the influences of roof slope on the characteristics of snow drifting on monoslope roofs during snowfall is still inadequate so far. Scaled tests of snow drifting on monoslope roofs with varying slopes (0° to 25°) were thus carried out in a wind tunnel equipped with a snowfall simulator, employing silica sand particles to model the snow precipitation environment. Typical snowdrift patterns on monoslope roofs were summarized, and elaborations were provided on the variation for the characteristics of snow drifting with roof slope. Similar features in terms of particle distributions, transport rate, and degree of erosion are observed for lower-sloped roofs (0°, 5°, and 10°), which all differ from those on higher-sloped roofs (15°, 20°, and 25°). The difference caused by the variation of approaching wind velocity on particle erosion is found to be more pronounced compared with that of snowfall intensity, and the increase in the degree of erosion with snowfall intensity can be reasonably represented by the linear relationship. The flow pattern around the roof and surface friction velocity before (without snow) and after snowfall (with snow cover) were then explored via numerical simulation for different roof slopes. Based on the simulated flow characteristics, the mechanisms of snow drifting on the sloped roofs under snowfall were further discussed. • Scaled tests of snow drifting on monoslope roofs with the concurrence of snowfall are conducted in a wind tunnel. • Effects of the approaching wind velocity and snowfall intensity on the characteristics of snow drifting are discussed. • Flow pattern around the roof and surface friction velocity before and after snowfall are explored via numerical simulation. • Typical snow distribution patterns on monoslope roofs with different slopes are presented. [ABSTRACT FROM AUTHOR]

In regions prone to heavy snowfall, an accurate estimation of snow redistribution on roofs under the action of wind is vital for structural engineers. The content of unbalanced snow loads of flat roofs caused by snow transport is indispensable in current load codes/standards. Hence, the wind tunnel tests were performed to investigate the redistribution of snow on flat roofs, in which high-density silica sand was used. The characteristics of snow redistribution on flat roofs are discussed, and common features are pointed out. The largest snow depth usually occurs near the windward region, and for a large-span flat roof the peak point could also occur in the rear region of the roof. In addition the locations of peak points after snow redistribution and the influence of wind velocity, wind duration and roof span on transport rates and mass fluxes are analyzed in detail. The transport rate increases as wind velocity or roof span increases but it is not a simple proportional relationship with roof span. Moreover, the transport efficiency of particles declines as wind duration becomes longer. Mass flux of the entire roof, which is the transport rate per unit length, asymptotically decreases with the increasing of roof span. [ABSTRACT FROM AUTHOR]

A Weather Research and Forecasting (WRF) model simulation with 3-km horizontal grid spacing was coupled with five California Meteorological Model (CALMET) simulations with horizontal grid spacing ranging from 100 m to 3 km to investigate the impact of horizontal resolution on the simulated near-surface wind fields of Super Typhoon Meranti (2016). Assessment of the model output was conducted in two specified regions named Shisun Mountain and Shishe Mountain in Fujian Province, China. The WRF/CALMET coupled system exhibited satisfactory performance, as indicated by the reasonably large correlation coefficient (>0.4) between the simulated and observed winds. The best and second-best outcomes for 10-m wind speed were found in the simulations with horizontal grid spacing of 500 and 250 m, respectively. Specifically, the 10-m wind speed appeared sensitive to model resolution after being affected by the gale-force winds of Typhoon Meranti. For the 10-m wind direction, the best outcome at Shisun Mountain was found with 500-m horizontal grid spacing, whereas it improved continuously with higher horizontal resolution at Shishe Mountain. The wind-followed slope angle was proposed to provide new perspective for quantitative evaluation of the impact of complex terrain with different horizontal resolutions on the near-surface wind field. It is found that, irrespective of the location and model resolution, the 10-m downhill winds were generally larger than the uphill winds before and after being affected by the gale-force winds, while the winds decreased with slope angle in both uphill and downhill situation when affected by the gale-force winds of Typhoon Meranti. • The performance of the WRF/CALMET coupled system was generally satisfactory. • The best and second-best results were found with grid spacing of 500 and 250 m. • A resolution of 100 m results in an overestimation of the measurements. • 10-m winds appeared sensitive to model resolution after being affected by typhoons. • The wind-followed "slope angle" is proposed for quantitative evaluation. [ABSTRACT FROM AUTHOR]