Comparison of different statistical methods used to estimate Weibull parameters for wind speed contribution in nearby an offshore site, Republic of Korea
The Weibull probability distribution indicates the probability of a specific wind speed and must be calculated before wind turbine installation. The Weibull distribution is affected by shape and scale parameters, which are driven in various ways. Many studies have conducted research to determine a m...
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | article |
| Lenguaje: | EN |
| Publicado: |
Elsevier
2021
|
| Materias: | |
| Acceso en línea: | https://doaj.org/article/f054633a512042e5a71a61bac03ade34 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | The Weibull probability distribution indicates the probability of a specific wind speed and must be calculated before wind turbine installation. The Weibull distribution is affected by shape and scale parameters, which are driven in various ways. Many studies have conducted research to determine a more reliable method among various Weibull parameter estimation methods. However, since these studies showed different results, studies on determining the higher reliable Weibull parameter estimation methods continues. In this study, we analyzed 10 years of data collected at the same location and height level in Maldo island(from 2010 to 2019) and Saemangeum seawall (from 2011 to 2012), the Republic of Korea. While former studies tried to rank the Weibull distribution methods based on the statistical analyses, in this study, we compared the Weibull parameters using twelve methods and identified the highest reliable and efficient methods for deriving the Weibull probability distribution by using the new approach comparing the variance of RMSE, R2, and χ2, which give a comprehensive insight about the level and fluctuations errors. These twelve methods are Alternative maximum likelihood method, Equivalent energy method, Empirical method of Justus, Empirical method of Lysen, Energy pattern factor method, Graphical method, Modified energy pattern factor method, Maximum likelihood method, Moment method, Modified maximum likelihood method, Power density method, Standard deviation method. The results showed while Empirical method of Justus, Empirical method of Lysen, Moment method, and Standard deviation method had the best accuracies in prediction of wind speed distribution, some methods such as Graphical method, Alternative maximum likelihood method, Equivalent energy method, and Energy pattern factor method had the worst prediction of wind speed distribution based on all variance of statistical methods for both regions. |
|---|