Statistical analysis of wind characteristics based on Weibull methods for estimation of power generation in Lithuania

Abstract The study analyzes application of Weibull probability distribution methodologies by summarizing wind power density in selected locations. Reliability of determination of Weibull probability density function of shape k and scale c parameters has been analyzed by means of eight methods. For the assessment of reliability of the methodology, root mean square error, coefficient of determination, chi-square test and relative error were calculated. Measurements of wind characteristics have been carried out in the coastal and continental part of the Lithuania. It has been determined that many calculation methods of the probability density function allow obtaining quite reliable results. However, depending on the geographical location of the area, the height from the ground level, and the influence of other factors on the wind power density, some methods are not acceptable, as they give over-large (up 13.44% and more) relative errors. A simulation model has been established to describe the wind power resources of locations with different climate conditions. For calculation Weibull parameters, it has been used 8760 of wind speed hourly data in 2014. The proposed model could be successfully used for the finding the suitable locations for development wind energy in regions which has the similar surrounding settings.