A study on wind energy generation forecasting using connectionist models

Wind generation is the most widespread form of renewable energy, with a high degree of penetration in traditional electricity networks. Hence, the influence of wind power generation over the efficient operation of power systems is increasingly complex. This determines the actors playing in the wind energy market to show an increased interest in developing efficient forecasting models for power generated in wind plants. This paper presents a study on wind energy generation forecast for a wind power plant located in the South - Eastern region of Romania, using a connectionist model. The forecasting model estimates the wind energy generation, based on weather forecasts for wind speed and direction. The model was designed based on an analysis conducted to determine the optimal structure of the Artificial Neural Network (ANN).