LAUSR

Abstract The distribution model of wind-speed data is critical for the assessment of wind-energy potential because it reduces uncertainties in the estimation of wind power output. Thus, an accurate distribution model for describing wind-speed data should be determined before a detailed analysis of energy potential is conducted. In this study, information from several goodness-of-fit criteria, e.g., the R2 coefficient, Kolmogorov–Smirnov statistic, Akaike’s information criterion, and deviation in skewness/kurtosis were integrated for the conclusive selection of the best-fit distribution model of wind-speed data. The proposed approach integrates standardized scores and subjects each criterion to multiplicative aggregation. The approach was applied in a case study to fit eight statistical distributions to hourly wind-speed data collected at two stations in Malaysia. The results showed that the proposed approach provides a good basis for the selection of the optimal wind-speed distribution model. Furthermore, graphical representations agreed with the analytical results.