SANTOS, S. A.; http://lattes.cnpq.br/8334809149422137; SANTOS, Samira de Azevedo.
Resumo:
The simultaneous use of optimized insolation data and multivariate statistical techniques was used in this work aiming to evaluate the relationship between the variations in rain rainfall pattern of Natal-RN and the availability of solar radiation as an alternative resource for electricity production. The research is exploratory and multivariate character. The exploratory nature is due to the fact it does not exist in the literature a validated model generation insolation data on hourly partition. The multivariate character from the need to investigate complex relationships between multiple atmospheric processes. The use of statistical techniques (factor analysis, cluster analysis and discriminant analysis) allowed to evaluate patterns of association between sunlight and other weather variables. It was possible to infer the influence of extreme weather conditions on the availability of solar radiation as a source of electricity generation. The results show that the atmospheric circulation (wind) plays an important control environmental conditions and consequently the energy transfer. The pattern of cloudiness and rain is a determining factor in the modulation of solar radiation that passes through the atmosphere and plays an important role availability of solar energy. However, it was found that regardless of the period analyzed (dry or rainy anomalous) convective nature of the precipitation in the region and with rains generally concentrated in a few days of the month ensures privileged environmental conditions of use of solar energy. Statistical data obtained in the calibration process and model validation through comparative analysis between calculated and observed insolation data, demonstrate good performance and reliability in the use of SPDH as an important tool in the development of solar energy utilization projects. However, results of statistical analyzes based on scatter measurement and the mean square error diagrams, it is clearly observed a tendency to underestimate the values. The biggest discrepancies, however, are associated mainly with data obtained in the early hours of the day and from 17 and 18 local time. These times correspond to times with lower rates of direct irradiance which implies softer burning heliogram.