Modelagem da curva de potência geral do Parque Eólico Sotavento

Abstract

The Sotavento Experimental Park in Galicia underwent a performance analysis of its wind turbines. By collecting wind speed and power data over time, it was possible to estimate the overall power curve of the park. The results showed that the actual performance is lower than expected, indicating lower efficiency in converting the kinetic energy of the wind into electrical energy. A polynomial regression curve analysis was conducted, revealing that higher-degree models fit the data better. However, beyond degree 6, there are no significant gains in the quality of the fit. This suggests that a polynomial higher than degree 10 is not necessary to accurately describe the data. The theoretical generation curve showed a maximum potential of 17.86 MW, but the maximum value observed in the actual curve was only 9.11 MW, representing approximately 51% of the maximum potential. The park was also capable of generating power at low wind speeds, with a minimum observed value of 0.29 MW. The park's capacity factor was calculated at 65.95%, indicating that it is capable of generating on average almost two-thirds of the possible energy if it were operating continuously at its maximum capacity. However, replacing outdated generators with more modern and efficient equipment could further improve the energy generation capacity. One of the identified limitations was the low efficiency of the technology used at certain points in the power curve. However, retrofitting with full-load converters is a viable option to increase efficiency and generation capacity without replacing all the turbines, reducing modernization costs. Modernization would bring benefits such as improved efficiency and generation capacity, reduced operational costs, and increased equipment lifespan. It is recommended to consider replacing outdated equipment with more efficient technologies to better harness the park's potential. Power curve analysis is crucial in assessing the generation capacity of a wind park. Furthermore, utilizing programming language algorithms such as Python, in platforms like Jupyter Notebook, can be an effective way to process and analyze large volumes of data, providing accurate and reliable results not only in power curve analysis but also in wind speed data analysis and wind power production forecasting.