Improving artificial neural network model for the prediction of the wind energy potential in Izmir Aliaga Region

In this study, it has been investigated that which parameters are mainly effected according to the meteorological conditions of the regions where the power generation capacities of wind turbines are installed and the models that can predict the power and energy production in the future according to these parameters can be realized by using artificial neural networks method. In this study, 5 input 1 output ANN model was developed to predict power generation based on Cp, U, T, A and P parameters. Model 0.99 was able to predict successful accuracy. Then, It was observed of these parameters, that Cp, U, T and P parameters of 4 combinations; Cp, U and T parameters of 3 combinations; Cp and U parameters of 2 combinations together and U parameter alone can make a more effective estimation. At the end of the research, a mathematical model and application program, which can make ANN estimation according to U parameter has been developed. In the study, wind energy estimation model has been developed by taking into account the wind speed of the region.