Improved wind resource modeling using bimodal Weibull distribution

Despite the common norm of modeling wind regimes using two-parameter Weibull distribution (2WD), this probability density function (PDF) is not suitable for sites of calm wind regime. Furthermore, 2WD PDF lacks bimodality and exhibits a quasi-flat shape, which are phenomena seen occasionally in some sites. In this paper, the application of bimodal Weibull distribution (BWD) is proposed as a more comprehensive alternative to the conventional 2WD. A comparative analysis of BWD with 2WD, five-parameter Weibull and Weibull distribution, and three-parameter generalized extreme value distribution, across 32 sites spanning all five continents, reveals moderate to substantial improvements in root mean square error, chi(2) statistic, and R-2. In addition, the paper demonstrates and explores distinct attributes of BWD, such as bimodality, quasi-flat shapes, flat-start, and others.