Journal of Fundamentals of Renewable Energy and Applications
Open Access
ISSN: 2090-4541
+44 1300 500008
ISSN: 2090-4541
+44 1300 500008
Bartolome Manobel Ponce, Ignacio Salfate Lara and Sonia Montecinos Geisse
Universidad de La Serena, Chile
Scientific Tracks Abstracts: J Fundam Renewable Energy Appl
The correct estimation of the power generated by a wind turbine is a key issue to the provision of the electricity that the wind farm will transfer to the grid and for a correct evaluation of the performance of each turbine. The aim of this study is to present a simple, practical model that can be easily implemented to calculate the power developed by each turbine in a wind farm. Considering the different atmospheric and operational variables involved and the non-linear relationships between those variables, we developed a model based on artificial neural networks. Input data, mainly atmospheric variables, can be directly measured in the park or be results of predictions from meteorological models. We took into consideration the influence of the operational parameters when modeling the turbine availability. This factor has never been included in calculations of the power generated by wind farms. The result improves widely-used models that include polynomial fitting for adjusting the power curve and those based on the standards of the International Electro-technical Commission. An adjustable model for each individual turbine is presented which results in prediction errors of wind power forecasting of only 5% of the root mean square error. This model can be used for the evaluation of the performance of a turbine itself and for forecasting the power generated by a wind farm using data from numerical weather models.
Bartolomé Manobel Ponce has a MEng in Industrial Engineering and a BEng in Mechanical Engineering from University of Seville, Spain. He is currently working on a research project on renewable energy at the University of La Serena, Chile.
Email: bmanobel@userena.cl