Wind distribution analysis incorporating Artificial Bee Colony Algorithm

Abstract

Wind speed distribution analysis is essential for assessment of the wind energy potential and also performance of wind energy conversion system. Two-parameter Weibull is the commonly used Probability density function (PDF) to model wind speed distribution. Conventionally method of maximum likelihood (MLE) and method of moments (MOM) methods are used for parameter estimation. In this paper Artificial Bee Colony (ABC) algorithm is applied to compute shape and scale parameters of Weibull distribution function. Statistical parameters such as maximum error in the Kolmogorov-Smirnov test and coefficient of determination (R2) are considered as judgment criteria to test the goodness of fit of the Probability density function. Results show that parameter estimation incorporating Artificial Bee Colony (ABC) algorithm is better than conventional iterative solving of MLE and MOM methods.