Modeling optimal PV surface of BIPVs for maximum energy yield through Genetic Algorithms

Abstract

Price decrease of photovoltaic (PV) energy in recent years allows to an increase in the use of small-scale PV modules, especially in the residential sector. Considering that houses and buildings are exposed to solar radiation during the day, it is possible to generate electricity by installing PV modules at strategic locations within the construction, using facades, windows, walls and roofs as Building Integrated Photovoltaic (BIPV) applications. Based on the above, it is analyzed the geometry of Colombian terraces corresponding to an area located on top of houses and buildings, which have flat open spaces that serve as roofs. Uses of terraces are related to drying clothes and cultivation; therefore, it is possible assign a portion of the entire area to the generation of electricity from the installation of PV modules. Occupation area due to integrated BIPV applications, is an important issue and is often limited to the flat geometry of most commercial solar panels. Thus, the goal of this work, is to generate curved photovoltaic surfaces that maximize the area exposed to the sun, seeking to increase the power of radiation captured and using the same area of occupation of conventional PV modules. To achieve this, a Genetic Algorithm (GA) is integrated into a computational ray tracing tool, in order to evaluate the radiation captured by a set of surfaces, which have a projected flat area on the terrace of 1m2, considering historical irradiance data in Medellín, Colombia. The optimal PV surface obtained, presents an additional 19.3% of captured energy and an additional area of 28% compared to the flat geometry. This is a promising finding because it shows a direct relationship between radiation capture and the geometric shape of the PV surface. The performance of the Optimal PV surface is validated and compared with flat geometry using TracePro.