Frequency Distribution of Wind Speed Near the Surface

Open Access
Publisher Website
Google Scholar
Abstract
The purpose of this paper is to provide information about frequency distributions of wind speed near the surface of the earth. There has been a recent increased interest in this topic because of the potential for obtaining electrical energy from wind power. Frequency distributions of wind speed from a large collection of data are examined. Most distributions are skewed to the right and the mean is usually greater than the median. Experience has shown that the Weibull distribution provides a good analytical approximation to the cumulative distribution and is particularly useful for the 90–99% thresholds. Two methods of fitting a Weibull distribution with a nonzero location parameter are discussed. Both of these methods require less computational effort than the maximum likelihood solution for a three-parameter model and are suitable for practical use by the engineer. It is shown that the three-parameter model is better than the two-parameter model for predicting extreme values. Abstract The purpose of this paper is to provide information about frequency distributions of wind speed near the surface of the earth. There has been a recent increased interest in this topic because of the potential for obtaining electrical energy from wind power. Frequency distributions of wind speed from a large collection of data are examined. Most distributions are skewed to the right and the mean is usually greater than the median. Experience has shown that the Weibull distribution provides a good analytical approximation to the cumulative distribution and is particularly useful for the 90–99% thresholds. Two methods of fitting a Weibull distribution with a nonzero location parameter are discussed. Both of these methods require less computational effort than the maximum likelihood solution for a three-parameter model and are suitable for practical use by the engineer. It is shown that the three-parameter model is better than the two-parameter model for predicting extreme values.

This publication has 0 references indexed in Scilit: