Inverse Burr distribution for extreme wind speed prediction: Genesis, identification and estimation

Abstract

The randomness of the wind source is a concerning issue for managing power plants in reliable conditions. High values of wind speed are undesirable since wind farms provide zero power for values greater than their cut-off thresholds. Also, the mechanical safety of the installations can be seriously compromised by extreme values of wind speed. Therefore, a reliable estimation of extreme values of wind speed is mandatory. An Inverse Burr distribution is proposed as an useful alternative for the probabilistic modeling of extreme values of wind speed. Distribution parameters were estimated through maximum likelihood and moment estimation procedures, and through a new proposal, the quantile estimation procedure. The proposed model is validated on several real wind datasets, comparing the proposed model with commonly-used extreme value models. Numerical applications showed that the proposed model is a valid and feasible alternative to the classical extreme value distributions for extreme values of wind speed.