Search

Diffuse solar irradiation (H D) data are essential for the design and management of photovoltaic solar systems, biosphere-atmosphere modeling, and other applications. However, H D observations are scarce in several locations, especially in tropical regions. Employing hourly diffuse solar irradiation (H D h) and global solar irradiation (H G h) data collected between 2002─2003 and 2007─2008 in Alagoas State, Northeast Brazil, this study assesses various modeling techniques. Empirical models, including third-degree polynomial, logistic, sigmoidal, and rational functions, were compared with AI methods such as artificial neural networks (ANN), support vector machine (SVM), and adaptive neuro-fuzzy inference system (ANFIS). Additionally, it explores how solarimetric and meteorological variables impact the performance of these models. The empirical models showed similar performance in estimating K D h (= H D h / H G h) (r2 > 0.726, modified Willmott – d m > 0.704, and RMSD < 0.103), with the third-degree polynomial model standing out. The empirical models had difficulty estimating K D h for hourly atmospheric transmittance (K T h) > 0.80, which indicated that they are not able to adequately simulate clear sky conditions, mostly due to surface reflections and clouds at the end of the day. ANN (r2 > 0.718, d m > 0.702, and RMSD < 0.105) showed better precision and accuracy of estimates for a greater number of schemes in relation to SVM and ANFIS (r2 > 0.704, d m > 0.699, RMSD < 0.108) and to empirical models. AI methods were able to represent the complexity of these conditions, with overall performance in estimating K D h superior or equivalent to empirical models. This study underscores the significance of exploring diverse methods for H D estimation, demonstrating promising potential for accurate and reliable estimation of hourly diffuse solar irradiation. • Hourly diffuse solar irradiation fraction (K D h) was estimated using atmospheric transmittance(K t h). • Empirical models and artificial intelligence (AI) methods were assessed. • Solarimetric and, or meteorological variables were considered on AI methods. • The empirical models and AI methods had similar overall performance. • The AI methos estimating better the complexity of K D h for K T h > 0.80 [ABSTRACT FROM AUTHOR]