Estimating mixture hybrid Weibull distribution parameters for wind energy application using Bayesian approach

The Weibull distribution function is essential for planning and designing wind-farm implementation projects wind-resource assessments. However, the limited those areas with high frequencies of calm winds. One solution to use hybrid distribution. In fact, when wind speed data present heterogeneous structures, it makes sense group them into classes that describe different regimes. single presents fitting errors should be minimized. this context, mixture distributions represent an appropriate alternative modelling wind-speed data. This approach was used combine associated by weighting contribution each them. study proposes based on mixtures in West Africa region. focused (WW-BAY) (WWH-BAY) using Bayes' theorem characterize over twelve years recorded at Abuja, Accra, Cotonou, Lome, Tambacounda sites Africa. parameters models were computed expectation-maximization (E-M) algorithm. estimated initial values provided Levenberg-Marquardt results obtained from proposed compared classical whose are some numerical method such as energy pattern factor, maximum likelihood, empirical Justus methods statistical criteria. It found WWH-BAY model gives best prediction power density Cotonou Lome sites, relative percentage error 0.00351 0.01084. factor lowest Abuja site a value -0.54752, Accra -0.55774, WW-BAY performs well 0.19232. recommended these useful applications African