A Dynamic Equivalence Method Considering the Spatial Effect of Wind Farms

In this paper, a dynamic and equivalent modeling method of large-scale wind farms based on clustering algorithm and measured data is proposed. According to the significant difference between wind speed and power curves in different wind turbine sets, the spatial effect is considered using the random sampling comparison method. In order to highlight the influence of spatial effect on the output power of wind farm, a simulation model consisting 20 wind turbines on the MATLAB/Simulink simulation platform is built. The result shows that the spatial effect cannot be neglected if the dynamic equivalent models of large-scale wind farm need a higher accuracy. In the actual wind farm, the measured wind speed and power data have to be taken into account the influence of the spatial effect. Therefore, the measured data of a wind farm is used as the clustering index according to K-means clustering analysis method. In an actual wind farm, 33 sets of UP77-1.5 MW wind turbines are grouped into 4 clusters. Each wind turbine set corresponds to an equivalent model, and takes into account the spatial effects of the sets. Finally, according to the comparison and error analysis of each model and the measured data, the rationality and correctness of the dynamic equivalent model proposed in this paper are verified. Compared with the traditional model, the model established in this paper is more accurate than the traditional model in the dynamic characteristics of the wind farm.