Evaluating wind power potential for a site is indispensable before making any decision for the installation of wind energy infrastructures and planning for relating projects. This paper presents a branch of a composite analysis whose objective was to investigate the potential of wind energy resource in Rwanda. Statistical methods were used to analyze long term time series of monthly daily wind speed measured on four meteorological stations in Rwanda for the period between 1981 and 1993. The Weibull distribution was used to model empirical distribution of measured long term monthly average wind speeds. The scale and shape parameters were estimated using four methods, the least square method, the likelihood method, the method of moments and the Chi-square method. It was observed that the method used for the estimates of the probability density of wind speed and giving the best overall fit to the distribution of the measured wind data varies from a location to another. However, the energy output calculated using wind speeds derived from the Chisquare method gave the best overall fit to the empirical distribution of the wind power density.
Keywords: Wind speed; Least square method; Likelihood method; Moments Method; Chi-square method; Weibull distribution; Wind energy; Energy; Rwanda.