Estimation methods for global solar radiation: Case study evaluation of five different approaches in central Spain

Résumé

Solar radiation can be estimated by a variety of methods in an attempt to overcome the limitations of on-ground records. Novel methods are often appearing but these are rarely compared to others from a different approach. This study surveys the main types of estimation methods for daily Global Horizontal Irradiation (GHI), and then, one characteristic technique per group is selected, discarding possible hybrid approaches: a parametric model based on temperatures and precipitation (Antonanzas model), a statistical model (XGBoost), interpolated ground-based measurements (Ordinary Kriging (OK)), a satellite-based dataset (CM-SAF-SARAH), and a reanalysis dataset (ERA-Interim). The techniques are evaluated in relation to the seasonal variation, the clearness index and the spatial performance at 38 grounds stations in central Spain from 2001 to 2013.Three different tiers of estimations were obtained being SARAH and OK the best performing methods overall. The SARAH dataset (MAE=1.10±0.13MJ/m2, MBE=0.22±0.36MJ/m2) generated estimates with the lowest spread, but led to a slight overestimation in low-altitude flat areas. The OK (MAE=1.10±0.25MJ/m2, MBE=0.00±0.31MJ/m2) outperformed SARAH in these flat areas (high density of stations), but at the expense of a higher variability. Alternatively, SARAH surpassed Ordinary Kriging (OK) when the distance to the closest station exceeded 20-30km. The ERA-Interim reanalysis and the XGBoost were in the second tier of estimations, and the parametric model yielded the worst results overall. ERA-Interim exhibited a systematic overestimation. The locally trained Antonanzas and XGBoost struggled to model the atmospheric transmissivity, showing large positive errors in spring months and a small underestimation of clear-sky days. Finally, a summary with the strengths and weaknesses of the five methods provides a deeper understanding for the selection of the adequate estimation approach. © 2016 Elsevier Ltd.