Digital numerical map-oriented estimation of solar energy potential for site selection of photovoltaic solar panels on national highway slopes

Abstract

Photovoltaic systems are promising replacements for fossil fuels at places where high solar energy is available. The estimation of available solar energy is the key to maximizing energy generation because sites with high available solar energy must be selected. Previous approaches to site selection required several experts to avoid subjective biases, often relying on rough estimations where the topography was not fully considered. Therefore, this study proposes a computational method that estimates the potential of solar energy for prioritizing and selecting sites for photovoltaic solar panels using publicly available digital numerical maps. These maps provide various kinds of spatial data to support proper land utilization. We focus on the use of national highway slopes as potential installation sites because they are typically unused public areas with high accessibility and few restrictions. For the estimation of annual solar irradiation, elevation contours and highway networks are first separated from the digital numerical maps. The extracted contours are subsequently used to generate a set of topographic data for the estimation. The extracted highway network is used to produce a binary mask representing fill slopes within which grid cells of high solar energy are identified and clustered to locate suitable sites. The proposed method is applied to a test site near Gongju, South Korea, where top 10 potential installation sites are clustered and ranked based on their estimated annual solar irradiation. The most suitable site is identified in eastern Gongju with an annual solar energy of approximately 49,623.7 MWh.