Abstract: A statistical technique to characterize wind speed data for use in wind energy conversion
system (WECS) is presented. We start by examining the frequency distribution of long-term data. The
histogram is generated for observed wind speed for a particular hour over a month for a number of
years. It is fitted to two distributions (Weibull, and β).Genetic Programming (GP) also used to fit the
distribution. Two goodness-of-fit criteria are employed in checking the best fit. These are Chi-square
(χ2), and Kolmogorov-Smirnov (KS). Long-term data from International Logan Airport, Boston, MA,
USA are analyzed. It is found that the GP fits the long-term hourly data best.