Evaluation of three methods for estimating annual and seasonal wind speed distributions

In this study annual and seasonal wind speed variations of M'sila (a region of northern Algeria) were examined based on data measured at an elevation of 10 m above the ground level during three years (2008-2010). Wind speed model is a critical factor in assessing wind energy potential and also in estimating the performance of wind energy conversion systems. In this paper, Weibull function, hybrid Weibull function and the probability function derived with maximum entropy principle (MEP) were used and compared with the measured data. To test their performance, statistical parameters such as correlation coefficient R², root mean square error (RMSE) and Chi square error (χ²) were considered. According to the results of failure analysis the hybrid Weibull probability distribution density function is the better alternative to conventional Weibull and MEP probability density functions, when describing annual and seasonal wind speed distributions in M'sila region.