Hierarchical Bayesian Model for Estimating Spatial-Temporal Photovoltaic Potential in Residential Areas

Abstract

This paper presents a Bayesian hierarchical model to estimate the spatial-temporal photovoltaic potential in residential areas. The proposed model offers a probabilistic approach that uses technical criteria of planners and favorable socioeconomic conditions for installing photovoltaic systems. Thus, the inhabitants' distrust of the photovoltaic solar energy choice is modeled via random distributions. The results are a spatial database that allows the creation of thematic maps to visualize the spatial distribution of photovoltaic potential in cities' residential areas for each year of the planning horizon. The proposed methodology was applied to a medium-sized city in Brazil. Maps which came from the application show the subareas with higher photovoltaic potential, where a range of impacts could appear on the distribution networks. Therefore, the results can contribute to multiscenario planning and operation studies of low- and medium-voltage networks performed by utility companies.