中文摘要：

针对现有方法无法细致全面地描述多维风电出力相关性的问题，创新性地提出结合改进K-means聚类、C藤结构和D藤结构建立混合藤Copula模型，以此描述多维风电出力相关性。以美国某地区相邻的3个风电场实测出力数据为例，在IEEE30节点系统中对所提方法进行验证。算例结果表明，依据混合藤Copula模型所获取的模拟数据能更准确地模拟多维风电出力的相关性，得到更精确的概率潮流计算结果。

英文摘要：

To solve the problem that the existing methods can＇ t detailedly and comprehensively describe the correlation between multidimensional wind farm output, a mixture vine Copula structure model combining the improved K-means clustering, C vine copula and I） vine Copula was proposed in this paper, through which a more accurate correlation model can be obtained. The validity of the mixture vine Copula structure model is tested in IEEE 30 node system with measured data of 3 adjacent wind farms in a certain area, and the results indicate that the correlation of orignal data can be expressed by the simulated data of the mixture vine copula structure more accurately, the power load flow results can be achieved more precise by the simulated data.