基于模糊C均值聚类的风电场多机等值方法

Multi-machine Equivalent Approach of Wind Farm Based on Fuzzy C-means Clustering

  • 摘要: 风电场等值是含风电场接入电网分析计算的重要技术手段。为降低风电场等值的难度,提高风电场分群的效率,本文基于风电机组实际运行中的监测状态量,采用模糊C均值(FCM)聚类算法,实现了风电场等值。首先选定各机组输出有功功率、无功功率、机端电压有效值及输出电流有效值为分群指标,并根据给定的等值机台数,将风电场分群问题转化为聚类问题;其次建立了风电机组类属隶属度函数和模糊C均值聚类算法的目标函数,通过迭代求解最优的聚类中心和模糊隶属度矩阵,得到风电场分群结果,算法具有计算简单、收敛性好的特点;然后,根据分群结果,对不同群的风电机组进行等值,实现风电场的多机等值;最后,通过仿真比较验证了本方法的有效性。本方法选取的分群指标具有可实操性,且在给定等值机台数条件下,计算更为简单、等值精度更高,适合用于风电场等值的实际工程计算。

     

    Abstract: The equivalent model is one of the most important mean to analyze the operation of power grid with wind farms. In order to reduce the difficulty of the equivalence of wind farm and to improve the efficiency of the grouping of wind farm, fuzzy C-means (FCM) algorithm is applied to equivalence of wind turbines in the wind farm based on the monitored operational states of wind turbine generator system (WTGS). Firstly, the output active and reactive power, voltage and current of WTGS are adopted as grouping index, and the wind farm grouping problem is transformed into a clustering problem according to the given number of equivalent WTGS. Secondly, the fuzzy membership function indicating WTGSs category and the objective function of FCM clustering algorithm are given. The clustering result is obtained based on the optimal cluster center and fuzzy membership matrix solved by iterative solution, and FCM is a simple algorithm with good convergence. Thirdly, according to the grouping result, the multi-machine equivalent model of the wind farm is achieved through the equivalence of different groups of WTGS. In the end, the validity of the approach is verified through simulation. The clustering index in FCM is more practical than other index. For given number of equivalent WTGS, the calculation of FCM is simpler and the accuracy is higher too. This method is very suitable for the equivalent model of wind farm in practice.

     

/

返回文章
返回