Statistical analysis ANOVA and MANOVA of irradiation and temperature database for photovoltaic systems

Photovoltaic solar energy has been exploited as a solution to meet the growing demand for electricity through a renewable source of energy. Optimized simulations of photovoltaic systems require the use of accurate solarimetric and climate data. This paper aims to compare the difference in solar radiation and temperature data provided by the NASA SSE and Meteonorm databases with the data measured by a real solarimetric station installed in Campinas, Brazil. Analyzes were performed using statistical methods such as analysis of variance from ANOVA and MANOVA. The analyzes present the hypothesis of equality with the significance of 95 % for each case is not rejected. Hypothesis rejection results occur in MANOVA analysis for temperature and irradiation data with daily data. The paper concludes that NASA SSE data is closest to the actual data.