基于数据驱动相似度方法的风电场站动态等值建模研究  

Research on dynamic equivalent modeling of a wind farm using a data‑driven degree of similarity method

作　　者：吴岳 朱林[1] 胡永浩 刘阳 WU Yue;ZHU Lin;HU Yonghao;LIU Yang(School of Electric Power,South China University of Technology,Guangzhou 510640,China)

机构地区：[1]华南理工大学电力学院,广东广州510640

出　　处：《电力科学与技术学报》2024年第5期118-128,共11页Journal of Electric Power Science And Technology

基　　金：国家自然科学基金(U22B6007)。

摘　　要：针对大型集中式风电场站中风电单元众多、运行工况多变、集电网及拓扑接线繁杂的特点,提出一种数据驱动相似度方法,实现对大型集中式风电场站的等值建模。首先,引入相似度来表征风电发电单元运行状态中的数据特征,并通过相似度,以数据驱动方式实现风电发电单元聚类分群;其次,对同一集群的发电单元进行聚合,获得等值机组的等值参数,并最终得到风电场等值模型;最后,以广东省某海上风电场为实例,对所提方法进行仿真与验证。研究结果表明,该方法能有效提升风电场的建模效率和模型精度。Considering the characteristics of numerous wind power units,variable operating conditions,and complex collection grids as well as topological wiring in large-scale centralized wind farms,a data-driven similarity method is proposed to realize the equivalent modeling of such wind farms.Firstly,similarity is introduced to characterize the data features in the operating states of wind power generation units,and through similarity,data-driven clustering of wind power generation units is achieved.Secondly,the generation units within the same cluster are aggregated to obtain the equivalent parameters of the equivalent units,ultimately leading to the equivalent model of the wind farm.Finally,a case study of an offshore wind farm is used to simulate and verify the proposed method.The research results indicate that this method can effectively enhance the modeling efficiency and accuracy of wind farms.

关 键 词：风电场 动态等值 数据驱动 物理量 相似度 谱聚类分群 

分 类 号：TM614[电气工程—电力系统及自动化]