Statistical distribution modeling of global solar radiation in Alagoas, Brazil: A comparative study (2008-2016)

• Global solar radiation in 6 cities (2008-2016) shows significant seasonal variation. • Highest solar radiation in Pão de Açúcar, lowest in São Luiz Quintude. • EV distribution is more accurate for Maceió’s unique radiation pattern. • This paper emphasizes statistical models' role in solar planning. This research investigates the global solar radiation levels in six cities of Alagoas, Brazil, from 2008 to 2016, assessing the suitability of various probability distributions to model solar radiation data. The study revealed that the highest global solar radiation values were observed in Pão de Açúcar, averaging 1600 kJ/m², while the lowest was in São Luiz Quintude, with a maximum of 3800 kJ/m² and a minimum of 16 kJ/m² for Maceió. The data, including hourly and seasonal variations, highlighted substantial fluctuations in solar radiation, particularly during the dry season when higher radiation levels were recorded. Several statistical distributions, such as Weibull, Generalized Extreme Value (GEV), Extreme Value (EV), and Lognormal, were evaluated to model the solar radiation patterns accurately. The results indicated that the GEV function best fits the data for most cities, with the EV distribution being more suitable for Maceió. The analysis further demonstrated that accurately modeling solar radiation using statistical distributions is crucial for implementing solar energy projects. The findings underscore the need for long-term observed data to account for variations and improve the prediction accuracy of solar radiation in the region.