Solar Radiation Decomposition Models in Alpine Environments: modeling tools and comparison with measurement.

When conducting a feasibility study for the installation of photovoltaic panels, it is crucial to know the direct and diffuse components of solar radiation that reach the ground. Numerous decomposition methods have been developed to estimate the diffuse part using different predictors because, in most cases, only the global component is available. However, these models have rarely been evaluated in mountainous environments. For this reason, the first objective of this study is to assess the performance of decomposition models in alpine contexts. Six models—Orgill & Hollands, Erbs, DISC, DIRINT, Reindl, and Boland—are assessed by comparing the estimated diffuse and beam irradiance with real measurements using datasets gathered from 19 Swiss meteorological stations at different elevations. The second goal is to formulate an approach for improving current models. The results suggest systematic biases in the models when applied at high altitudes, with the DIRINT model showing the best overall performance. In addition, the study demonstrates that a re-calibrated model can lead to significant improvements under certain conditions. These findings provide valuable guidance for refining decomposition models in mountainous areas and contribute to advancing the energy transition through more effective use of renewable energy sources.