A Novel Algorithm to Estimate Solar Irradiance of Urban Buildings for Photovoltaic Potential Estimation System Using a 3D City Model

Abstract. Photovoltaic (PV) power generation is one of most promising means to prevent global warming in the present Japan. Since solar panels mounted on building façades are expected to come into wide use in urban areas, accurate estimation of PV potential of building façades is necessary for urban energy management planning. Accordingly, we decided to develop a system to estimate PV potential of urban buildings using a 3D city model. The system has two significant features: rapid estimation of hourly solar irradiance of points densely distributed on a building surface, and more flexible estimation of PV potential considering an arrangement of solar panels on a building surface. The paper reports our newly developed algorithm to estimate solar irradiance of urban buildings. The algorithm adopts the idea that “whether the sun can see us or not” indicates “whether we have sunshine or not” for calculation of hourly solar irradiance of a point on a building surface. In estimation of solar irradiance distributions on roofs and façades, utilization of projection images viewed from the sun created by using computer graphics (CG) techniques such as the depth buffer (Z-buffer) algorithm makes our system have much less computation time than most of existing systems using a hemispherical viewsheds or ray tracing. Results of an experiment conducted in Yokohama of Japan demonstrate that the algorithm would be able to estimate solar irradiance on not only roofs but also façades of urban buildings using a 3D city model accurately enough.