凌佳凯,章逸舟,胡金峰,秦军,戴健,费有蝶,朱振
LING Jiakai,ZHANG Yizhou,HU Jinfeng,QIN Jun,DAI Jian,FEI Youdie,ZHU Zhen

• 1:国网江苏省电力有限公司无锡供电分公司
• 2:河海大学电气与动力工程学院

摘要(Abstract)：

配电网中准确的拓扑结构辨识对运行和控制具有重要意义，针对实际配电网拓扑结构变动的情况，搭建了可智能辨识配电网拓扑结构的深度学习模型。首先，生成不同拓扑结构下的配电网量测数据并进行数据预处理。其次，构建了融合CNN（卷积神经网络）、LSTM（长短期记忆网络）和Attention（注意力机制）的拓扑结构智能辨识模型，并结合历史量测数据对模型训练并测试。最后，在IEEE 33节点和PG&E69节点配电系统仿真算例中，验证了该基于CNN-LSTM-Attention模型的拓扑辨识方法相较于传统辨识方法在辨识精度上的优越性，实现了该模型的在线应用。
Accurate identification of the topology in a distribution network is crucial for its operation and control. Addressing the dynamic changes in the actual topology of distribution networks, an intelligent deep learning model capable of recognizing distribution network topologies was developed. Firstly, measurement data for distribution networks under different topologies were generated, followed by data preprocessing. Subsequently, an intelligent topology identification model was constructed, integrating convolutional neural network(CNN), long short-term memory network(LSTM), and Attention mechanism. The model was trained and tested using historical measurement data.Finally, in simulation scenarios using the IEEE 33-node and PG&E69-node distribution systems, the superiority of this CNN-LSTM-Attention-based topology identification method over traditional approaches in terms of identification accuracy was validated, and online application of the model was achieved.

关键词(KeyWords)： 配电网;拓扑辨识;卷积神经网络;长短期记忆网络;注意力机制
distribution networks;topology identification;convolutional neural network;long short-term memory network;Attention mechanism

Abstract:

Keywords:

基金项目(Foundation): 国网江苏省电力有限公司科技项目（J2021026）

作者(Author): 凌佳凯,章逸舟,胡金峰,秦军,戴健,费有蝶,朱振
LING Jiakai,ZHANG Yizhou,HU Jinfeng,QIN Jun,DAI Jian,FEI Youdie,ZHU Zhen

DOI: 10.19585/j.zjdl.202403010

参考文献(References)：

• [1]马钊，张恒旭，赵浩然，等.双碳目标下配用电系统的新使命和新挑战[J].中国电机工程学报，2022,42(19):6931-6945.MA Zhao,ZHANG Hengxu,ZHAO Haoran,et al.New mission and challenge of power distribution and consumption system under dual-carbon target[J].Proceedings of the CSEE,2022,42(19):6931-6945.
• [2]韩平平，张楠，潘薇，等.配电网状态估计可观性研究综述[J].电力系统及其自动化学报，2022,34(4):11-21.HAN Pingping,ZHANG Nan,PAN Wei,et al.Review of research on observability of distribution network state estimation[J].Proceedings of the CSU-EPSA,2022,34(4):11-21.
• [3]唐广瑜，金鑫琨.基于全网供电拓扑模型的图模数据共享技术研究与应用[J].电测与仪表，2022,59(6):105-112.TANG Guangyu,JIN Xinkun.Research and application of graphic model data sharing technology based on power supply topology model of the whole network[J].Electrical Measurement&Instrumentation,2022,59(6):105-112.
• [4]高骞，杨俊义，刘凯，等.基于数据驱动技术的配电网拓扑结构及线路参数识别方法[J].电力电容器与无功补偿，2023,44(2):77-87.GAO (Jian Qian),YANG Junyi,LIU Kai,et al.Identification method of distribution network topology and line parameters based on data driven technology[J].Power Capacitor&Reactive Power Compensation,2023,44(2):77-87.
• [5]陈沛东，曹华珍，何璇，等.中压配电网近邻交互式分布式拓扑辨识算法[J].电力工程技术，2023,42(2):139-146.CHEN Peidong,CAO Huazhen,HE Xuan,et al.Distributed topology identification algorithm of medium-voltage distribution network based on neighboring interaction[J].Electric Power Engineering Technology,2023,42(2):139-146.
• [6]张秋瑞，何柏娜，王银忠，等.低压配电网拓扑异动自适应识别与校验[J].山东电力技术，2022,49(8):28-34.ZHANG Qiurui,HE (Bai Bo)(Na),WANG Yinzhong,et al.Adaptive identification and verification of low voltage distribution network topology anomalies[J].Shandong Electric Power,2022,49(8):28-34.
• [7]杨金东，李文.基于K-Means算法的配电网负荷和拓扑等效方法[J].电力电容器与无功补偿，2022,43(2):147-153.YANG Jindong,LI Wen.Equivalent method of distribution network load and topology based on K-means algorithm[J].Power Capacitor&Reactive Power Compensation,2022,43(2):147-153.
• [8]ZHANG J W,WANG Y,WENG Y,et al.Topology identification and line parameter estimation for non-PMU distribution network:a numerical method[J].IEEE Transactions on Smart Grid,2020,11(5):4440-4453.
• [9]王果，郭文凯，王长春.配电网拓扑辨识研究综述及展望[J].吉林大学学报（工学版），2023,53(2):312-327.WANG Guo,GUO Wenkai,WANG Changchun.Overview and prospect of distribution network topology identification[J].Journal of Jilin University (Engineering and Technology Edition),2023,53(2):312-327.
• [10]MILI L,STEENO G,DOBRACA F,et al.A robust estimation method for topology error identification[J].IEEETransactions on Power Systems,1999,14(4):1469-1476.
• [11]LUGTU R,HACKETT D,LIU K,et al.Power system state estimation:detection of topological errors[J].IEEETransactions on Power Apparatus and Systems,1980,99(6):2406-2412.
• [12]刘家齐，刘浩军，童力，等.低压配电物联网台区拓扑识别技术测试研究[J].浙江电力，2022,41(12):12-20.LIU Jiaqi,LIU Haojun,TONG Li,et al.Testing and research on topology identification technology for Io T-type low voltage distribution area[J].Zhejiang Electric Power,2022,41(12):12-20.
• [13]陶华，杨震，张民，等.基于深度优先搜索算法的电力系统生成树的实现方法[J].电网技术，2010,34(2):120-124.TAO Hua,YANG Zhen,ZHANG Min,et al.A depthfirst search algorithm based implementation approach of spanning tree in power system[J].Power System Technology,2010,34(2):120-124.
• [14]ZHAO J A,LI L A,XU Z,et al.Full-scale distribution system topology identification using Markov random field[J].IEEE Transactions on Smart Grid,2020,11(6):4714-4726.
• [15]ZHAO Y E,CHEN J S,POOR H V.Efficient neural network architecture for topology identification in smart grid[C]//2016 IEEE Global Conference on Signal and Information Processing (Global SIP),December 7-9,2016,Washington DC,USA:[s.n.],811-815.
• [16]童力，梁海维，邹旭东，等.基于数据驱动的配电网拓扑识别及线路阻抗估计[J].浙江电力，2022,41(1):10-18.TONG Li,LIANG Haiwei,ZOU Xudong,et al.Datedriven distribution network topology identification and line impedance estimation[J].Zhejiang Electric Power,2022,41(1):10-18.
• [17]孙伟，朱世睿，杨建平，等.基于图卷积网络的微电网拓扑辨识[J].电力系统自动化，2022,46(5):71-81.SUN Wei,ZHU Shirui,YANG Jianping,et al.Topology identification of microgrid based on graph convolutional network[J].Automation of Electric Power Systems,2022,46(5):71-81.
• [18]GOTTI D,AMARIS H,LEDESMA P.A deep neural network approach for online topology identification in state estimation[J].IEEE Transactions on Power Systems,2021,36(6):5824-5833.
• [19]裴宇婷，秦超，余贻鑫.基于Light GBM和DNN的智能配电网在线拓扑辨识[J].天津大学学报（自然科学与工程技术版），2020,53(9):939-950.PEI Yuting,QIN Chao,YU Yixin.Online topology identification for smart distribution grids based on LightGBMand deep neural networks[J].Journal of Tianjin University(Science and Technology),2020,53(9):939-950.
• [20]LECUN Y,BOTTOU L,BENGIO Y,et al.Gradientbased learning applied to document recognition[J].Proceedings of the IEEE,1998,86(11):2278-2324.
• [21]YANG F F,SONG X B,XU F,et al.State-of-charge estimation of lithium-ion batteries via long short-term memory network[J].IEEE Access,2019,7:53792-53799.
• [22]VASWANI A,SHAZEER N,PARMAR N,et al.Attention is all You need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems,December 4-9,2017,Long Beach,USA:[s.n.],6000-6010.
• [23]BARAN M E,WU F F.Network reconfiguration in distribution systems for loss reduction and load balancing[J].IEEE Transactions on Power Delivery,1989,4(2):1401-1407.
• [24]章博，刘晟源，林振智，等.高比例新能源下考虑需求侧响应和智能软开关的配电网重构[J].电力系统自动化，2021,45(8):86-94.ZHANG Bo,LIU Shengyuan,LIN Zhenzhi,et al.Distribution network reconfiguration with high penetration of renewable energy considering demand response and soft open point[J].Automation of Electric Power Systems,2021,45(8):86-94.
• [25]BARAN M E,WU F F.Optimal capacitor placement on radial distribution systems[J].IEEE Transactions on Power Delivery,1989,4(1):725-734.
• [26]徐俊俊，吴在军，周力，等.考虑分布式电源不确定性的配电网鲁棒动态重构[J].中国电机工程学报，2018,38(16):4715-4725.XU Junjun,WU Zaijun,ZHOU Li,et al.Robust dynamic reconfiguration for distribution networks considering uncertainty of distributed generations[J].Proceedings of the CSEE,2018,38(16):4715-4725.