Site Selection of Large Ground-Mounted Photovoltaic Plants: A GIS Decision Support System and an Application to Italy

Abstract

Latterly, central governments and local authorities have been establishing various constraints on the construction of new large ground-mounted photovoltaic (PV) plants, because of the soil consumption, landscape impact, and also competitiveness with the crop production. This is particularly important in contexts where the agricultural sector is closely linked to the territory. With the aim of providing a decision support tool based on quantitative indicators for the site selection of large ground-mounted PV plants, in this article the criteria for the identification of areas suitable for the installation of ground-mounted photovoltaic systems, recently emerged by regional government or in the technical and scientific literature, are applied to the entire territory of the Piedmont region (25,000 km2). Both qualitative criteria for inclusion/exclusion (e.g., exclusion from areas of great value) and criteria for quantification (e.g., solar resource availability) were considered. The aggregation of the quantitative criteria into the final indicator is done by means of an Artificial Neural Network (ANN) trained with values corresponding to sites of existing PV plants in the Region. It emerges that the available areas are very limited, concentrated, and strongly influenced by the criteria of exclusion/inclusion. Some considerations on the significance of the results for the region of analysis are finally made.