Modelling wind speed dependence in system reliability assessment using copulas

Abstract

Owing to geographic relationships of wind farms, wind speeds at different wind sites show certain dependence between them, which may create a considerable impact on the reliability of power systems containing wind energy. A new technique for establishing multivariate distribution of wind speeds using copulas is presented to model wind speed dependence in the study. The separation between marginal distributions of each single wind speed and dependence structure is a basic feature of the copula method, which leads to the effectiveness and simplicity in dealing with the dependence in a multivariate distribution. The case studies using two actual wind speed data indicate that the copula model can give a valid representation of multiple wind speed series. The impact of wind speed dependence between two wind regimes on system reliability is illustrated using a commonly recognised reliability test system. The results confirm that system reliability will decrease as the wind speed dependence increases.