Identification of Suitable Probability Density Function for Wind Speed Profiles in Power System Studies

摘要

The uncertainty in the prediction of wind power generation can be reduced by accurately identifying the probability distribution for modelling wind speed, which is the most influential parameter in calculating wind power. This paper aims to identify the most appropriate probability density function (PDF) for wind speed to use in power system studies, particularly for low wind speed, where the most commonly used Weibull distribution unable to produce a satisfying representation. Therefore, this paper has tested various probability density functions (PDF), which include Rayleigh, Weibull, Gamma, Lognormal, Normal, Inverse Gaussian, Generalized Extreme Value and Exponential distributions, to identify an accurate PDF for modelling low wind speed data. Root mean square error (RMSE) and the coefficient of determination (R^{2) are used as the measure of accuracy. The obtained results have indicated that the Gamma distribution (followed by Generalized Extreme Value distribution) provides the best representation for modelling wind speed data. This result has been further verified by performing probabilistic power flow simulation in the IEEE-30 bus test system, where the accuracy of the voltage profile PDF follows the same trend similar to wind profile PDF.}