Statistical scrutiny of Weibull parameters for wind energy potential appraisal in the area of northern Ethiopia

Abstract

Paramount two-parameter Weibull function has been extensively used to assess the wind energy potential. The performance contrast of four statistical methods, i.e., energy pattern factor method, least squares regression method, method of moments and mean standard deviation method in estimating extensively used Weibull parameters for wind energy application at four selected locations of northern Ethiopia has been studied. The contrast of statistical methods is compared through relative percentage error, root mean square error, mean percentage error, mean absolute percentage error, Chi-square error and analysis of variance (or) efficiency of the methods used. Test results evidently revealed that, least squares regression method presents better performance than other methods selected in the investigation. The least efficient methods to fit the Weibull distribution curves for the assessment of wind speed data especially for four selected locations are energy pattern factor method, method of moments and mean standard deviation. From the actual data analysis, it is found that if wind speed distribution matches well with the Weibull function, the above three methods are applicable, but if not, least squares regression method can be considered based on the cross checks including energy potential and cumulative distribution function.