Use of MinMaxEnt distributions defined on basis of MaxEnt method in wind power study

Abstract

Knowledge of the wind speed distribution is an important information needed in evaluation of wind power potential. Several statistical distributions have been used to study wind data. The Weibull distribution is the most popular due to its ability to fit most accurately the variety of wind speed data measured at different geographical locations throughout the world. Recently, maximum entropy (MaxEnt) distributions based on the maximum entropy method have been widely used to determine wind speed distribution. Li and Li used the MaxEnt distribution for the first time in the wind energy field and proposed a theoretical approach to deter-mine the distribution of wind speed data analytically. Ramirez and Carta discussed the use of wind probability distributions derived from the maximum entropy principle in the analysis of wind energy. In this study, MinMaxEnt distributions defined on the basis of the MaxEnt method are introduced and applied to find wind distribution and wind power density. A comparison of the MinMaxEnt and Weibull distributions on wind speed data taken from different sources and measured in various regions is conducted. The wind power densities of the considered regions obtained from the Weibull and MinMaxEnt distributions are also compared. The results indicate that the MinMaxEnt distributions obtained show better results than the known Weibull distribution for wind speed distributions and wind power density. Therefore, MinMaxEnt distributions can be used to estimate wind distributions and wind power potential