Copula Graphical Models for Wind Resource Estimation

Kalyan Veeramachaneni, Alfredo Cuesta-Infante, Una-May O'Reilly

Abstract: We develop multivariate copulas for modeling multiple joint distributions of wind speeds at a wind farm site and neighboring wind source. A n- dimensional Gaussian copula and multiple copula graphical models enhance the quality of the prediction site distribution. The models, in comparison to multiple regression, achieve higher accuracy and lower cost because they require less sensing data.