考虑多风电场相关性的场景概率潮流计算及无功优化  

Scenario probabilistic load flow calculation and reactive power optimization considering wind farms correlation

作　　者：邱宜彬 欧阳誉波 李奇[1] 陈维荣[1] 

机构地区：[1]西南交通大学电气工程学院,四川成都610031 [2]国网湖南省电力公司张家界供电分公司,湖南张家界427000

出　　处：《电力系统保护与控制》2017年第2期61-68,共8页Power System Protection and Control

基　　金：国家科技支撑计划(2014BAG08B01);国家自然科学基金(51177138;61473238;51407146);四川省杰出青年基金(2015JQ0016)~~

摘　　要：针对结合K-means聚类和Copula函数建立场景概率模型时,K-means聚类不能根据风电出力数据分布特点自发确定最佳聚类数这一不足,提出通过基于密度的聚类有效性指标确定最佳聚类数。并以此建立最优场景概率模型,采用改进型回溯搜索算法(BSA)进行无功优化。以澳大利亚的两个相邻风电场实测出力为例,在含多风电场的IEEE30节点系统中对所提方法进行验证,算例结果表明采用所提方法确定的最优场景概率模型能准确描述多风电场输出功率之间的相关性。When combining K-means clustering with Copula theory to build scenario-based probabilistic model,K-means clustering can＇t spontaneously determine the optimal number of clustering.For this deficiency,this paper proposes a density-based index to determine the optimal number of clustering.Then the optimal scenario-based probabilistic model is built,based on which reactive power optimization is conducted by adopting modified backtracking search algorithm（BSA）.At last,the proposed method is verified in IEEE30 system with data measured from two adjacent wind farms in Australia,and simulation results indicate that adopting optimal number of clustering determined by proposed method could get an accurate description of correlation between wind farms.

关 键 词：场景概率潮流 无功优化 K-means最佳聚类数 COPULA函数 改进型回溯搜索算法 

分 类 号：TM614[电气工程—电力系统及自动化] TM744