Akademik Veri Yönetim Sistemi
2nd International Conference on Pure and Applied Mathematics (ICPAM-VAN 2018), Van, Türkiye, 11 - 13 Eylül 2018, ss.73
Precisely and efficiently modeling wind speed is important in estimating the wind energy potential
of a specified region. Two- parameter Weibull distribution is the most widely used and accepted
distribution in the energy literature. However, it does not model the all wind speed data encountered
in nature. Therefore, in this study, different distributions are used for modeling wind energy, such
as Gamma, log- normal, Rayleigh. The estimators of the unknown parameters of these distributions
are found by using maximum likelihood estimators (MLEs). Finally, hourly wind speed data (m/s),
measured hourly at 10m from Adilcevaz and anakkale Turkey during December 2017 is used. These
data are taken from the Turkish State Meteorological Service. The fit distribution is determined with
respect to root mean square error (RMSE), Kolmogorov Smirnov and coefficient of determination
R2
criteria