Investigating the use of readily accessible climatic data for predicting vertical solar irradiance under sunlit and shaded scenarios

For evaluating active and passive energy systems, vertical global irradiance (G_VT) data are required. However, unlike horizontal irradiance, G_VT measurement is sparingly available, making G_VT models an alternative. Also, previous GVT models have been developed using diffuse irradiance (D_VT). Nevertheless, estimation methods for determining the D_VT are a huge source of computational error in G_VT modelling. This study proposed G_VT models for two G_VT cases (i.e., sunlit and shaded vertical surfaces) based on the vertical direct and reflected irradiance. By separating these cases into two, the error-prone D_VT estimation method was omitted. The input variables used were combinations of the ratio of direct normal irradiance to horizontal global irradiance (D_NI/G_HI), clearness index (K_t) and scattering angle (χ). The method was improved by proposing simple regression and machine learning (ML) models for G_VT prediction. ML was also used for feature importance identification. All models were developed using 10 min of measured data from Hong Kong. Findings show that D_NI/G_HI and K_t are important variables for modelling G_VT for the sunlit and shaded surface, respectively. Also, aside from the linear regression models, the ML models had %RMSE ranging from 8.9 to 13.1 % and 12.8–20.3 % when tested against the 2005 and 2019 to 2020 data, respectively. With most predictions having %RMSE less than 20 %, all models (especially the optimised support vector) gave good predictions of G_VT. Overall, the proposed models will be useful to building designers and researchers in deriving irradiance data for building energy evaluation. © 2025 Elsevier Ltd.