Open AccessWind-ice Joint Probability Distribution Analysis based on Copula Function
Author(s) -
Fuchun Yang,
Hongjie Zhang,
Qi Zhou,
Shanshan Liu
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1570/1/012078
Subject(s) - weibull distribution , wind speed , return period , copula (linguistics) , gumbel distribution , joint probability distribution , probability distribution , cumulative distribution function , goodness of fit , statistics , probability density function , marginal distribution , meteorology , mathematics , extreme value theory , econometrics , physics , geography , random variable , archaeology , flood myth
Taking the wind speed and ice thickness field measurement data in Southwest China during November 2016 to March 2018 as analytical sample, the probability distribution of the wind speed and ice intensity were analyzed and fitted by using Gumbel distribution model, Weibull distribution model and Generalized extreme value distribution (GEV) model, respectively. The goodness of fitting results had been compared. After the determination of marginal distribution functions for wind speed and ice intensity, 500-year pseudo wind speed and ice intensity samples were generated based on the Monto Carlo method. Five Copula functions were employed in the construction of joint probability distribution function and the most suitable Copula function was chosen for building the wind-ice joint probability distribution model. On the basis of the model, the return periods for 30 years, 50 years and 100 years had been calculated, respectively. The analysis results show that the GEV model well matches the characteristics of wind speed and ice intensity. As for binary Copula function, the Frank Copula function has the best goodness for the joint probability distribution of wind speed and ice intensity. The return period calculations show that the design value of wind-ice joint return period (JRP) is larger than that of co-occurrence return period (CRP) and the value of Kendall return period (KRP) is between that of JRP and CRP. Considering that the dangerous areas of return periods are different, the return period for design should be determined by the combination analysis of different kinds of the return periods. Meanwhile, the importance and the failure results of the project should be taken into account.