PROBABILITY DISTRIBUTIONS OF WIND SPEED IN THE CAMPO GRANDE, MS, BRAZIL

The objective of this study is to model wind characteristics using twelve probability density functions: Weibull, Ralyeigh, Log-Logistics, Inverse Gaussian, Normal, Range, Extremely Generated, Extreme, Lognormal, Logistic, Burr and Rician. Four statistical criteria, coefficient of determination, mean square error, mean absolute error and mean absolute error are considered as judgment criteria to evaluate the adequacy of probability density functions. As a result, Weibull, Rayleigh, generalized extreme value, extreme value, and Rician distributions accurately perform data. These distributions can be used as an alternative distribution that adequately describes wind speed data in Campo Grande. The weaker settings are obtained by the Normal, Burr, Logistics, Log-Logistic, and Inverse Gaussian distributions