Abstract: The clustering of wind turbines is of important significance for improving the accuracy of equivalent wind farm modeling. Therefore, the advantages and disadvantages of the conventional methods were analyzed, based on which, a clustering method according to the dynamical characteristic curves of doubly-fed wind turbines (DFIG) was then proposed. By using this method, the similarities of dynamical characteristic curves of DFIGs under faulty condition are firstly calculated to be the cluster index. Based the calculation, the hierarchical clustering method is used to cluster the wind turbines. Secondly, wind turbine operations can be segmented with the proposed clustering method, and the segmentation way is verified by the simulation. In the final section, the comparisons among different clustering methods are carried out. The result has demonstrated the feasibility of the proposed method.

Key words: wind farm with DFIG, clustering method, dynamical curve, operation segmentation, equivalent modeling