摘要
随着电网规模的不断扩大,电网联系越来越密集,短路电流水平日益增高。220k V电网分区运行已成为限制500/220k V电磁环网短路电流的重要措施之一。为此,提出一种改进GN(Girven Newman)分裂算法的分区方法,该方法在以节点收缩后的节点重要度择取特征节点集成员的基础上,进一步采用GN分裂算法获取评价线路薄弱程度的边介数,而得到分区方案;同时为合理评估方案的优劣,在构建二级指标信息熵评价模型和一级综合指标主成分分析评价模型的基础上,提出一种综合考虑短路电流、静态安全约束及其社团结构模块度3类指标,基于信息熵主成分分析的环网分区方案评估方法。最后以新英格兰10机39节点系统和某实际电网作为算例,验证了所提分区方法的正确性和方案评估方法的有效性。
With the continuous expansion of grid, the connection of grid is more and more intensive. At the same time, short circuit current level is rising. The partition of 220kV grid layer has become an important measure in order to limit more short-circuit current. In this connection, the paper put forward a kind of electromagnetic loop network partition method based on improved GN （Girven Newman） algorithm. The method evaluated the importance of each node by applying network condensation degree quantitatively. Furthermore, traditional GN division algorithm partition would be employed to make splitting schemes in order to improve efficiency of partition. At the same time, the paper proposed a method of 220kV loop network partition schemes combining information entropy and principal component analysis, which emphasize three indicators those are short circuit, static-state security analysis and modularity of community structure for assessment separately. There were entropy evaluation model of the secondary index and principal component analysis and evaluation model of comprehensive indicator of level one. Finally, the article verified effectiveness of the algorithm and correctness of assessment method through New England 10-unit 39-bus system and an actual grid.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2017年第6期1686-1694,共9页
Proceedings of the CSEE
基金
国网四川省电力科学研究院科技项目(SGSCDK OOXTJS1500196)~~
关键词
电磁环网分区
节点重要度
GN分裂算法
短路电流
信息熵
主成分分析法
electromagnetic loop network partition
nodeimportant degree
GN splitting algorithm
short circuit current
information entropy
principal component analysis method
