Construção de um protótipo para simulação dos ângulos da geometria solar e correção da radiação incidente para quaisquer inclinações e desvios azimutais

Abstract

Solar power generation systems have shown great promise, especially as they are renewable energy sources. However, the use of this energy in residential photovoltaic installations is not fully utilized when the modules are installed on roofs without any inclination correction. Although there are some mathematical equations in the literature that make it possible to accurately calculate the angles of solar geometry and perform corrections for solar radiation, many designers dispense with this procedure due to the complexity of these equations. This work aims to build a prototype that is capable of proving the mathematical equations that describe solar geometry and carrying out the simulation of all solar movement throughout the year, thus making it possible to visualize in a practical and didactic way the angles of solar geometry, and to carry out the correction of incident solar radiation for different inclinations and for an anisotropic sky. The prototype was developed at the State University of Western Paraná, and has a mechanical structure capable of simulating solar positioning following a mathematical formulation presented by several authors. The system is controlled by a microcontroller that locates the exact position of the solar azimuth and zenith angles, corresponding to the desired solar position for any day and time, between sunrise and sunset. Statistical quality control tools were used to analyze the results, which aim to measure the accuracy, reliability and relevance of the data. The standard literature limits of ±3σ were considered as a tolerable variability, referring to each interaction analyzed. The result based on the errors obtained between the expected values and the observed values revealed consistent and precise patterns in several measurements, both in the aspect related to the zenith, azimuthal and incidence angles, with coefficients of determination (R2) very close to 1, indicating that the observed value is highly precise in relation to the expected value, and the process has great reproducibility capacity.