Pesquisa sobre o efeito de fenômenos solares no potencial energético solar-eólico

Abstract

This thesis analyzes the monthly data from historical series of the heat stroke cycles, solar radiation and average wind speeds in the 1961 and 2008 period to identify long-term inaccuracies in the location of both wind and solar sources. The state of Rio Grande do Sul-Brazil, was chosen as a case study, to estimate the behavioral trend of these variables and compare them with the measured data, testing the homogeneity of information. Therefore, it aims at improving the long-term prognosis in locating projects of solar power generation plants. The analysis of these climatic variables was carried out using ARIMA models (autoregressive integrated moving average models) as well as the Box & Jenkins methodology and seasonality studies with the X11 ARIMA models with 5% statistical significance. In this study, the period between 1961 and 2011 indicated that heat stroke rates were not enough to overcome the values recorded in the 1960s and 1970s, when the percentages were in most months 1% below the historical average. The observed heat stroke data suggest decreasing trends in the 1980s and 1990s, due to the presence of the phenomenon called "global dimming", which contributed to lower levels of solar radiation.!A possible structural break has been found in the wind series from August 2001 through the CUSUMQ test (cumulative sum of squares of recursive waste) and the Lane et al test. (2002), leading to higher values and overestimating the final prognosis of wind power. A decrease in the average wind speed was also observed from 2003 to 2011 during six months of these years.!The spring season, often with the highest wind potential had the highest mean decrease while the season with the lowest wind potential, fall, had the opposite behavior during the studied period. By spectral analysis, performed by Fourier method, the time series of sunshine and solar radiation also showed cycles with possible ranges of influence on measurements of the solar energy potential. Such temporal variations in the data, indicate that possible locations for the wind and photovoltaic plants may be seriously affected since longterm weather fluctuations can vary significantly even at the best location selected to generate electricity.