Assessment of Wind Energy Resources Using Data Mining Techniques

Abstract

The paper presents the application of Data Mining techniques for the assessment of wind energy resources at Lodz Hills (Wzniesienia Lodzkie). The measurements taken at a meteorological station, as well as long-term data from meteorological reanalysis were used as the input data. Linear regression, neural networks and support vector networks were used to obtain a long-term forecast of potential wind energy resources. According to the European and Polish recommendations specifying the suitability of land for wind turbines installation in terms of wind conditions, the obtained forecast confirmed the purposefulness of localization of such installations in the examined area. The purposefulness of applying Data Mining methods for solving problems related to the assessment of wind energy resources was also confirmed.