Estimating low-occurrence wind speeds from mean velocity and turbulent kinetic energy: Development of statistical method and validation with idealized cases

A novel statistical method (namely KB method) for estimating low-occurrence strong wind speeds (LOSWSs) from mean wind velocity and turbulent kinetic energy (TKE) was developed in this study. An isolated building case and a building array case from a large-eddy simulation (LES) database were used for the statistical analysis and validation. The statistics including the coefficient of variation (CV), peak factor (PF) K_q and gust factor (GF) G_q were analyzed. It was found that the relationships of the CV to the PF and GF of the LES results are close to the theoretical relationships of the two-parameter Weibull distribution (2W). The LOSWSs s_q with q = 10%, 1% and 0.1% have positive correlations with the mean wind speed ⟨s⟩ and have negative correlations with G_q. The KB method was proposed based on 2W. The shape parameter β of 2W was estimated as an empirical function of the TKE ratio k/(K+k), where K and k are the mean and turbulent kinetic energy, respectively. Then, the estimated β was used for calculating the theoretical gust factor of 2W. Finally, the LOSWS was calculated from the estimated gust factor and mean wind speed. By using the mean wind velocity and TKE from the LES database, the KB method showed high accuracy for estimating the LOSWSs with q = 10%, 1% and 0.1%. Relative errors were within 20% at most points.