کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
292845 | 511081 | 2016 | 14 صفحه PDF | دانلود رایگان |
• The characteristics of dynamic pressures and turbulence effects on a saddle type roof are examined.
• The moment-based Hermit model does not work well for strong non-Gaussian pressures in flow separation region.
• The turbulence intensity and integral length scale have noticeable influences on the pressure coefficients.
• A quasi-steady method is examined for compensating the effect of low-frequency turbulence.
This study investigates the characteristics of dynamic pressures on a saddle type roof under normal and corner attacking angles in various boundary layer turbulent flows. Six turbulent flows are simulated in wind tunnel. Long-term wind pressure coefficients data in each flow condition are collected to determine the statistics of pressure coefficients including mean, standard deviation (STD), skewness, kurtosis, negative peak, peak factor and probability distributions. The characteristics of dynamic pressures and the turbulence effects are examined. The pressures in the flow separation region show non-Gaussian characteristics with highly skewed unimodal distributions or bimodal distributions. The widely used moment-based Hermit model approach has limitations in estimating the extreme value distributions and peak factors of strongly non-Gaussian pressure coefficients. The roof pressure coefficients are not sensitive to the mean wind speed profile. On the other hand, the turbulence intensity and integral length scale have noticeable influences on the pressure coefficients. As the low-frequency turbulence cannot be well generated by wind tunnel in general, this study explores a quasi-steady method for compensating the effect of low-frequency turbulence. The STD pressure coefficient in the flow-separation region caused by low-frequency turbulence can be compensated well when the conical vortex is not strong.
Journal: Journal of Wind Engineering and Industrial Aerodynamics - Volume 150, March 2016, Pages 1–14