A platform for research: civil engineering, architecture and urbanism
Interference effects on tail characteristics of extreme pressure value distributions
AbstractThis study investigates the interference effects on the tail characteristics of extreme pressure coefficients. Systematic wind tunnel tests are conducted, from which a large amount of data is generated to describe the tail characteristics of extreme pressure coefficients under interference effects. Through the identification of shape parameter of Generalized Extreme Value distribution (GEVD) function and the estimation of fluctuation level of extreme pressures, optimal peak pressure coefficients are predicted to compare with the Cook and Mayne coefficients. Results show that the design fractiles based on estimated parameters are generally lower than 78%, and the identified concave tail characteristics imply a ceiling coefficient value. The largest negative peak pressure coefficients are found when the interfering building is in the upstream region of the principal building and most of them are along the vertical edges of side faces parallel to the wind direction. By calculating the error percentages due to the assumption of Gumbel tail, the estimation bias is indicated and the significances of both the shape parameter and the fluctuation level of extreme pressures are presented.
HighlightsLarge ensemble size of pressure data are fitted to find shape parameters, coefficient of variation under interferences.An iteration procedure to estimate design fractile and optimal peak pressure coefficient is demonstrated.Differences between estimations based on Gumbel tail and non-Gumbel tail are discussed.Specific pressure locations are selected for discussions on significant errors from conforming Gumbel assumption.
Interference effects on tail characteristics of extreme pressure value distributions
AbstractThis study investigates the interference effects on the tail characteristics of extreme pressure coefficients. Systematic wind tunnel tests are conducted, from which a large amount of data is generated to describe the tail characteristics of extreme pressure coefficients under interference effects. Through the identification of shape parameter of Generalized Extreme Value distribution (GEVD) function and the estimation of fluctuation level of extreme pressures, optimal peak pressure coefficients are predicted to compare with the Cook and Mayne coefficients. Results show that the design fractiles based on estimated parameters are generally lower than 78%, and the identified concave tail characteristics imply a ceiling coefficient value. The largest negative peak pressure coefficients are found when the interfering building is in the upstream region of the principal building and most of them are along the vertical edges of side faces parallel to the wind direction. By calculating the error percentages due to the assumption of Gumbel tail, the estimation bias is indicated and the significances of both the shape parameter and the fluctuation level of extreme pressures are presented.
HighlightsLarge ensemble size of pressure data are fitted to find shape parameters, coefficient of variation under interferences.An iteration procedure to estimate design fractile and optimal peak pressure coefficient is demonstrated.Differences between estimations based on Gumbel tail and non-Gumbel tail are discussed.Specific pressure locations are selected for discussions on significant errors from conforming Gumbel assumption.
Interference effects on tail characteristics of extreme pressure value distributions
Lo, Yuan-Lung (author) / Tseng, Yu-Fan (author)
Journal of Wind Engineering and Industrial Aerodynamics ; 170 ; 28-45
2017-08-11
18 pages
Article (Journal)
Electronic Resource
English
Interference effects on tail characteristics of extreme pressure value distributions
Online Contents | 2017
|Extreme value distributions for peak pressure and load coefficients
Online Contents | 2008
|Extreme value distributions for peak pressure and load coefficients
Online Contents | 2008
|Extreme value distributions for peak pressure and load coefficients
British Library Conference Proceedings | 2008
|Extreme value distributions for peak pressure and load coefficients
Elsevier | 2007
|