Abstract:

Wind energy is one of the most widely used renewable energy systems in the world. Accurate planning of production of wind farms requiring high investment costs is of great importance. In this study, wind power estimation is made for different types of wind turbines discussed by using one year measured wind speed data. Wind speed data are measured at a height of 90 m and recorded at 10 minute intervals for 12 months. Six different wind turbines are used in the study: Gamesa G97, Suzlon S.88, Siemens SWT2.3, Nordex N100, Enercon E82 and Vestas V117. These estimations that will enable the producers to make feasibility have been realized by using artificial neural networks. As an artificial neural network, feed forward back propagation network with high convergence rate, which can produce successful results in nonlinear problem situations, is preferred. With the model created using artificial neural networks, wind speeds of twelve months are estimated respectively. In the performance analysis of the artificial neural network model, the regression values are obtained close to 1. As a result of this study, Siemens Swt-2.3 MW turbine is determined as the most suitable turbine for the region and it is determined as Gamesa G97-2 MW, Nordex 100-2.5MW, Enercon 82-3 MW Vestas 117-3.3 MW, Suzlon 88-2.1 MW turbines.

Keywords: