Inverse Weibull Method Application to wind Speed Modeling in Campo Grande-Ms Brazil

Abstract

Wind potential estimation requires an analysis of wind characteristics (wind speed density and wind direction). In this study, the applicability of two distribution models named Weibull and Inverse Weibull aiming to characterize the wind speed distribution in Campo Grande-Ms (Brazil) is investigated. The wind speed data collected from Campo Grande-Ms National Institute of Meteorology (INMET) at 10 m height for 5 years from January 2013 to December 2017, at an hour interval, are used. The method of maximum likelihood estimation is applied to calculate the parameters of the selected distributions. The best distribution function is chosen based on three goodness-of-fit statistics, namely; mean absolute percentage error (MAPE), root mean square error (RMSE), and coefficient of determination (R²). The obtained results indicate that the Weibull distribution provides a more accurate and efficient estimation than Inverse Weibull distribution. Therefore, Weibull distribution can be used to better estimate wind speed distribution in Campo Grande-Ms (Brazil) than Inverse Weibull distribution.