Run Duration Analysis of Surface Wind Speeds for Wind Energy Application

Arden B. Sigl Department of Civil Engineering, South Dakota State University, Brookings 57006

Search for other papers by Arden B. Sigl in
Current site
Google Scholar
PubMed
Close
,
Ross B. Corotis Department of Civil Engineering, Northwestern University, Evanston, IL 60201

Search for other papers by Ross B. Corotis in
Current site
Google Scholar
PubMed
Close
, and
Danny J. Won Sargent and Lundy Engineers, Chicago, IL 60603

Search for other papers by Danny J. Won in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

Hourly wind speed records are used to develop a model for the probability distribution of wind speed persistence above and below fixed reference speeds. Examination of duration histograms from 19 sites for records varying from 5–24 years leads to the development of a simple composite distribution. Enforcement of smooth behavior and a parameter sensitivity analysis allow the model to be interpreted in terms of a single free parameter, which is then shown to be highly correlated to the seasonal mean wind speed at a site.

Comparison of run duration results for 1 and 3 h data (the latter being the standard digitization presently used by the National Climatic Center) indicates a definite bias with the 3 h records. A correction scheme is derived to improve the 3 h results.

Abstract

Hourly wind speed records are used to develop a model for the probability distribution of wind speed persistence above and below fixed reference speeds. Examination of duration histograms from 19 sites for records varying from 5–24 years leads to the development of a simple composite distribution. Enforcement of smooth behavior and a parameter sensitivity analysis allow the model to be interpreted in terms of a single free parameter, which is then shown to be highly correlated to the seasonal mean wind speed at a site.

Comparison of run duration results for 1 and 3 h data (the latter being the standard digitization presently used by the National Climatic Center) indicates a definite bias with the 3 h records. A correction scheme is derived to improve the 3 h results.

Save