Wind Speed Distribution Modeling in the Greater Area of Chania, Greece

Abstract

The current study presents a wind speed regional assessment for the greater area of a Mediterranean coastal valley in the island of Crete, Greece. Wind speed and direction experimental data are employed from six sites, appropriately located to incorporate the effect of the main topographical features. The mean wind speed and direction assessment is performed for the overall and seasonal periods and distinct wind speed patterns are identified. The wind power density is estimated for each site and regarding the development of wind energy applications, two areas with different characteristics are established. The Weibull, Rayleigh, Lognormal, Gamma, and Inverse Gaussian distributions are assessed for their ability to model the experimental wind speed frequency distributions for the monthly and overall period. Their goodness-of-fit is assessed using the coefficient of determination and the χ² hypothesis test. Additionally, the visual inspection of their fits to the corresponding histograms is done and the error on the estimated mean wind speed and its variance is examined. The Gamma probability distribution function is proposed as an alternative to the Weibull distribution for the area under study.

Keywords:

• Wind speed
• Wind energy
• Wind speed distribution
• Wind power
• Weibull distribution

ACKNOWLEDGMENT

The authors would like to thank Dr. George Sgouros for his help during the analysis of the results of this work.