考虑空间相关性的风电场等值方法

Equivalent Model of Wind Farms Considering Spatial Correlation

由于大规模风电场的复杂性,对风电场的每台风电机组建立详细的模型将大大增加仿真的计算量,因此需要对风电场进行等值化简。在分析了风电场风电机组之间尾流效应、海拔高度、空间遮挡和机组停运等相互作用的基础上,推导出能够表征风电机组空间相关性的相关性系数,并以此为分群指标运用K-means聚类算法对风电机组进行分群。K-means聚类算法需要事先给出聚类个数,针对此问题对K-means聚类方法进行了改进,提出了有效性指标来确定最优分群数,并给出了等值模型的参数计算方法,得到了风电场的多机等值模型。通过算例仿真表明,该方法具有较高的精度,对K-means方法的改进也具有一定的有效性。

With the complexity of large-scale wind farm,modeling for each turbine in wind farm will lead to excessive computing. Wind farm need to be simplified. This paper infers the correlation coefficient which can reflect spatial correlation between wind turbines on the basis of study on wake effect, altitude, shelter of and outage of turbine. The correlation coefficient is used as clustering index of K-means clustering algorithm. Cluster number need to be given for K-means clustering algorithm. This paper proposes an effectiveness indicator to determine the number of the optimal clustering and gives the parameter calculating method to get a wind farm model represented by multi equivalent wind turbines. The simulation results show that the proposed method has high precision and the improvement of K-means clustering algorithm is effective.