Improving Long-Term Wind Speed Assessment using Joint Probability Functions Applied to Three Wind Data Sets

An algorithm for long-term wind speed correction is proposed that uses two reference wind data sets to correct the short-term measurements at the target site; the assessment is computed using one joint probability distribution function which is derivided from the three concurrent records in the short-term period and other two ones built from the concurrent reference data collected in both long-term and short-term periods. Validation has been carried out against Jpwind algorithm, which is based on the standard correlation analysis between two sites, showing comparable performance in one test case when high level of correlation is present between pairs of met-masts. Other tests, opportunely arranged using artificial wind data time series, show that the implemented method achieves better score, proving that the addition of one reference wind data set can potential increase the accuracy of the long-term assessment.