Probabilistic load flow analysis for power systems with multi-correlated wind sources

This paper presents a new probabilistic load flow algorithm which takes consideration of multi-correlated wind sources in the power network. The paper first uses linear approximation to obtain the injected power distribution of wind farm by considering cut-in and cut-out wind speed. Then a 5-point discrete distribution is deduced by the point estimation method. Based on the copula method, a bivariate model which can model the correlation between the wind farms is also developed. The spatiotemporal dependencies of two wind farms are analyzed and discussed. In order to model the high dimensional joint distribution, a combination method, which combines all possible bivariate distributions, is introduced to reduce the dimensional of multivariate distribution for the situation of more than two correlated wind farms. The proposed algorithms have been tested using the IEEE118 bus test system. The results indicate that the proposed algorithm can actually capture the probabilistic characteristics of the power systems with multi-correlated wind sources.