基于关系图卷积神经网络的新能源配电台区拓扑识别方法

谢宏; 张华赢; 梁晓锐; 陈煜; 杨林立; 周斌

谢宏,张华赢,梁晓锐,陈煜,杨林立,周斌.基于关系图卷积神经网络的新能源配电台区拓扑识别方法[J].电测与仪表,2024,61(7):94-102.
Xie Hong,Zhang Huaying,Liang Xiaorui,Chen Yu,Yang Linli,Zhou Bin.A topology identification method based on relational-graph convolutional network for distribution substation area with high renewables[J].Electrical Measurement & Instrumentation,2024,61(7):94-102.

基于关系图卷积神经网络的新能源配电台区拓扑识别方法

A topology identification method based on relational-graph convolutional network for distribution substation area with high renewables

DOI：10.19753/j.issn1001-1390.2024.07.014

中文关键词: 配电台区分布式光伏线户关系识别拓扑生成图神经网络

英文关键词:distribution substation area, distributed photovoltaic, line-user relationship identification, topology generation, graph neural network

基金项目:南方电网公司科技项目（090000KK52222133/SZKJXM20222115）

作者中文名	作者英文名	单位
谢宏	Xie Hong	南方电网公司新型智慧城市高品质供电联合实验室深圳供电局有限公司
张华赢	Zhang Huaying	南方电网公司新型智慧城市高品质供电联合实验室深圳供电局有限公司
梁晓锐	Liang Xiaorui	南方电网公司新型智慧城市高品质供电联合实验室深圳供电局有限公司
陈煜	Chen Yu	湖南大学电气与信息工程学院
杨林立	Yang Linli	湖南大学电气与信息工程学院
周斌	Zhou Bin	湖南大学电气与信息工程学院

摘要点击次数: 464

中文摘要:

针对传统拓扑识别方法难以适应高比例分布式光伏接入下低压配电台区电气耦合特性复杂的问题,本文提出了一种基于关系图卷积神经网络的新能源配电台区拓扑识别方法。首先,分析了分布式光伏接入对低压台区线户关系识别的影响机理,提出了高渗透率分布式光伏接入下配电台区的自适应线户关系识别方法,通过电压皮尔逊相关系数矩阵建模和全局自适应聚类方法实现线户关系识别。其次,基于新能源台区拓扑关联特性将配网节点关联分类匹配为分隔、上下、并列及光伏节点接纳关系,建立了适应分布式光伏接入的台区拓扑邻接矩阵模型。最后,提出了基于关系图卷积神经网络的配电台区拓扑生成算法,通过提取电压量测数据形成台区节点特征矩阵,基于关系图链接预测挖掘潜在节点关联关系逐步生成配电台区拓扑。算例仿真对比验证了所提拓扑识别方法的有效性,与传统算法相比可提升识别准确率4.3%以上。

英文摘要:

A proposed topology identification method, based on a relational-graph convolutional network, addresses the challenges faced by traditional methods in adapting to the complex electrical coupling characteristics of low-voltage distribution substation areas with a high proportion of distributed photovoltaic (PV) integration. Firstly, the paper analyzes the influence mechanism of distributed PV integration on the identification of line-user relationships in low-voltage substation areas, and presents an adaptive method for identifying line-user relationships in distribution substation areas with high-penetration distributed PV integration. This method achieves line-user relationship identification through voltage Pearson correlation coefficient matrix modeling and global adaptive clustering. Secondly, considering the topology association characteristics of distribution substation areas with high renewables, the paper classifies and matches distribution network node associations into separated, hierarchical, parallel, and PV node acceptance relationships. This establishes an adjacent matrix model of substation area topology that is specifically adapted to distributed PV integration. Finally, the paper proposes a distribution substation area topology generation algorithm based on a relational-graph convolutional network. By extracting voltage measurement data to create node feature portraits for the substation area and employing graph link prediction to uncover potential node association relationships, the distribution substation area topology is progressively generated. A case simulation comparison validates the effectiveness of the proposed topology identification method, which improves identification accuracy by more than 4.3% when compared to traditional algorithms.

查看全文查看/发表评论下载PDF阅读器

关闭