Evaluation of diverse methods used to estimate Weibull parameters for wind speed in various Algerian stations

Abstract

In this paper, seven numerical methods have been used to estimate Weibull parameters using wind speed data for various Algerian stations. The investigated methods are least square method (LSM), the maximum likelihood method (MLM), the empirical method (EM), the moment method (MOM), the modified maximum likelihood method (MMLM), the energy pattern factor method (EPFM) and empirical method of Iysen (EML). The results obtained from these methods have been compared with the provided data to find their accuracy and precision based on three statistical tests namely: the root mean square errors (RMSE), the correlation coefficient (R2) and the Chi-square test (χ2). Wind speed data used in the present study have been measured at six (06) Algerian weather stations. These data were recorded at 10 m above the ground during the period from 1981 to 2004. Results in this paper indicate that modified maximum likelihood method present the best fit to Weibull distribution and moment method provides the lowest degree of fit for all the stations. This study gives also a detailed analysis of wind characteristics in the study area. It appears from the results that Tindouf station is the windiest site especially in winter months.