Evaluation of wind speed regime variability and model fit performance using multi-component Weibull mixture distributions

Abstract

Accurate modelling of wind speed distributions is essential for reliable wind resources assessment and effective turbine siting decisions. This study evaluates three mixture Weibull formulations which include Bimodal Mixture Weibull (BMW), Modified Bimodal Mixture Weibull (MBMW) and Four-Component Mixture Weibull (FCMW) using twenty-one (21) years wind speed data from Abeokuta and Lagos in Southwestern Nigeria. Maximum Likelihood Estimation (MLE) was used to estimate the wind speed data while Goodness-Of-Fit (GOF) metrics and Monte Carlo validation determined each model’s ability to represent complex wind regimes. The MBMW improves the characterisation of transitional features with good performance both in low and high windspeed locations through visual inspection and metrics. MBMW’s RMSE in Lagos is 0.1119 lower than 2.4927 of BMW. The MBMW produced more accurate histogram matches than FCMW. The FCMW provides the most comprehensive representation of multimodal patterns, particularly in the coastal environment except for poor fit in Abeokuta, the area with low windspeed. These findings demonstrate the value of flexible mixture models for strengthening wind resource assessment in Nigeria and supporting broader renewable energy planning across West Africa.