欧锋,罗醒华,龙经纬,徐超群,赖国清,杨慧敏
OU Feng,LUO Xinghua,LONG Jingwei,XU Chaoqun,LAI Guoqing,YANG Huimin

• 1:广东电网有限责任公司云浮供电局
• 2:东南大学

摘要(Abstract)：

为进一步提高低压配电网户变关系和相位识别的准确性，提出一种基于图转换和迁移学习的低压配电网户变关系和相位识别方法。首先，引入了基于格拉姆角场的图转换方法实现用电数据的二维化，以更好地发现一维时序用电数据间的差异性；然后，针对低压配电网用户数据稀缺、获取途径有限、样本数量较少等问题，基于迁移学习利用预训练好的参数权重，构建了适合户变关系和相位识别的深度学习模型。通过实验验证，所提模型在户变关系识别和相位识别中的准确率较主流方法均有所提升，拥有更好的稳定性。
A method based on graph transformation and transfer learning is proposed to further enhance the accuracy of household-transformer relationships and phase identification in low-voltage distribution networks. Firstly, a graph transformation method based on Gramian angular field(GAF) is introduced to convert electricity consumption data into a two-dimensional representation, facilitating the identification of differences in one-dimensional time-series electricity consumption data. Next, to address challenges such as sparse user data in low-voltage distribution networks, limited data acquisition methods, and a scarcity of samples, a deep learning model suitable for householdtransformer relationship and phase identification is constructed using transfer learning and leveraging pre-trained parameter weights. Experimental validation demonstrates that the proposed model outperforms mainstream methods in both household-transformer relationship and phase identification, exhibiting improved accuracy and stability.

关键词(KeyWords)： 低压配电网;户变关系和相位识别;格拉姆角和场;迁移学习;深度学习模型
low-voltage distribution networks;household-transformer relationship and phase recognition;Gramian angular summation field;transfer learning;deep learning model

Abstract:

Keywords:

基金项目(Foundation): 南方电网公司重点科技项目035300KK52190077(GDKJXM20198298)

作者(Author): 欧锋,罗醒华,龙经纬,徐超群,赖国清,杨慧敏
OU Feng,LUO Xinghua,LONG Jingwei,XU Chaoqun,LAI Guoqing,YANG Huimin

DOI: 10.19585/j.zjdl.202403011

参考文献(References)：

• [1]赵普.基于量测数据的配电网拓扑与参数识别研究[D].长沙：湖南大学，2019.ZHAO Pu.Research on distribution network topology and parameter identification based on measurement data[D].Changsha:Hunan University,2019.
• [2]方延文，王鲁剑，王少敏.城市配电网现状及发展趋势探讨[J].中国电力教育，2010（增刊1):15-16.FANG Yanwen,WANG Lujian,WANG Shaomin.Discussion on the present situation and development trend of urban distribution network[J].China Electric Power Education,2010(S1):15-16.
• [3]陆俭国，何瑞华，陈德桂.中国电气工程大典-第11卷-配电工程[M].北京：中国电力出版社，2009.
• [4]ZHAO C X,SUN M H,HANYU.Distribution networks distribution network condition based maintenance application present situation and development[C]//2014 China International Conference on Electricity Distribution(CICED).September 23-26,2014.Shenzhen,China.IEEE,2014.
• [5]张勇军，刘斯亮，江金群，等.低压智能配电网技术研究综述[J].广东电力，2019,32(1):1-12.ZHANG Yongjun,LIU Siliang,JIANG Jinqun,et al.Research review on low-voltage intelligent distribution network technology[J].Guangdong Electric Power,2019,32(1):1-12.
• [6]李琮琮，范学忠，王清，等.基于用电信息采集系统的配电网台区识别[J].电测与仪表，2019,56(24):109-114.LI Congcong,FAN Xuezhong,WANG Qing,et al.Station area recognition of distribution network based on electricity information acquisition system[J].Electrical Measurement&Instrumentation,2019,56(24):109-114.
• [7]李亚.配电网台区用户智能识别系统研究[D].上海：上海电力学院，2017.LI Ya.Research on the intelligent transformer area recongnition system[D].Shanghai:Shanghai University of Electric Power,2017.
• [8]PAPPU S J,BHATT N,PASUMARTHY R,et al.Identifying topology of low voltage distribution networks based on smart meter data[J].IEEE Transactions on Smart Grid,2018,9(5):5113-5122.
• [9]李熊，王伟峰，葛玉磊，等.基于特征电流的户变关系识别方法[J].电测与仪表，2021,58(9):115-121.LI Xiong,WANG Weifeng,GE Yulei,et al.Research on user-transformer relation identification method based on characteristic current[J].Electrical Measurement&Instrumentation,2021,58(9):115-121.
• [10]LISOWSKI M,MASNICKI R,MINDYKOWSKI J.PLC-enabled low voltage distribution network topology monitoring[J].IEEE Transactions on Smart Grid,2019,10(6):6436-6448.
• [11]张丽强，丛伟，董罡，等.基于多元线性回归的单相电表相别判断方法[J].电力自动化设备，2020,40(5):144-156.ZHANG Liqiang,CONG Wei,DONG Gang,et al.Method for single-phase electric meter phase identification based on multiple linear regression[J].Electric Power Automation Equipment,2020,40(5):144-156.
• [12]连子宽，姚力，刘晟源，等.基于t-SNE降维和BIRCH聚类的单相用户相位及表箱辨识[J].电力系统自动化，2020,44(8):176-184.LIAN Zikuan,YAO Li,LIU Shengyuan,et al.Phase and meter box identification for single-phase users based on tSNE dimension reduction and BIRCH clustering[J].Automation of Electric Power Systems,2020,44(8):176-184.
• [13]刘苏，黄纯，侯帅帅，等.基于DDTW距离与DBSCAN算法的户变关系识别方法[J].电力系统自动化，2021,45(18):71-77.LIU Su,HUANG Chun,HOU Shuaishuai,et al.Identification method for household-transformer relationship based on derivative dynamic time warping distance and densitybased spatial clustering of application with noise algorithm[J].Automation of Electric Power Systems,2021,45(18):71-77.
• [14]杨涛，孙志达，唐明，等.基于Tensor Flow框架的改进BP户变关系识别方法[J].浙江电力，2021,40(8):25-32.YANG Tao,SUN Zhida,TANG Ming,et al.An improved BP method for transformer-user relationship identification based on Tensor Flow framework[J].Zhejiang Electric Power,2021,40(8):25-32.
• [15]DEKA D,BACKHAUS S,CHERTKOV M.Structure learning in power distribution networks[J].IEEE Transactions on Control of Network Systems,2018,5(3):1061-1074.
• [16]CAVRARO G,KEKATOS V,VEERAMACHANENIS.Voltage analytics for power distribution network topology verification[J].IEEE Transactions on Smart Grid,2019,10(1):1058-1067.
• [17]林佳颖，栾文鹏，余贻鑫，等.AMI量测用于配电网在线状态估计的可信度建模及分析[J].电网技术，2018,42(4):1191-1200.LIN Jiaying,LUAN Wenpeng,YU Yixin,et al.Credibility modelling and analysis of AMI measurements for on-line distribution state estimation[J].Power System Technology,2018,42(4):1191-1200.
• [18]LUAN W P,PENG J,MARAS M,et al.Smart meter data analytics for distribution network connectivity verification[J].IEEE Transactions on Smart Grid,2015,6(4):1964-1971.
• [19]YANG Z C,SHEN Y,YANG F,et al.Topology identification method of low voltage distribution network based on data association analysis[C]//2020 5th Asia Conference on Power and Electrical Engineering (ACPEE).June4-7,2020.Chengdu,China:IEEE,2020.
• [20]谷海彤，张远亮，卢翔智，等.基于深度学习的户变拓扑关系的识别方法研究[J].信息系统工程，2020(3):150-151.GU Haitong,ZHANG Yuanliang,LU Xiangzhi,et al.Research on recognition method of topological relationship of household variables based on deep learning[J].China CIONews,2020(3):150-151.
• [21]WANG Z,OATES T.Encoding time series as images for visual inspection and classification using tiled convolutional neural networks[C]//Workshops at the Twentyninth AAAI Conference on Artificial Intelligence,2015.
• [22]程文傲，徐明，高金峰.小电流接地系统单相接地故障选线空间域图像生成及融合方法[J].电力自动化设备，2021,41(7):97-103.CHENG Wenao,XU Ming,GAO Jinfeng.Spatial domain image generation and fusion method of single-phase grounding fault line selection for small current grounding system[J].Electric Power Automation Equipment,2021,41(7):97-103.
• [23]SEZER O B,OZBAYOGLU A M.Algorithmic financial trading with deep convolutional neural networks:time series to image conversion approach[J].Applied Soft Computing,2018,70:525-538.
• [24]SILVA D F,DE SOUZA V M A,BATISTA G E A PA.Time series classification using compression distance of recurrence plots[C]//2013 IEEE 13th International Conference on Data Mining.December 7-10,2013.Dallas,TX,USA:IEEE,2013.
• [25]WANG Z G,OATES T.Imaging time-series to improve classification and imputation[EB/OL].2015:ar Xiv:1506.00327.https：//arxiv.org/abs/1506.00327.pdf.
• [26]HOWARD A G,ZHU M L,CHEN B,et al.Mobile Nets:efficient convolutional neural networks for mobile vision applications[EB/OL].2017:ar Xiv:1704.04861.https：//arxiv.org/abs/1704.04861.pdf.
• [27]高昆仑，杨帅，刘思言，等.基于一维卷积神经网络的电力系统暂态稳定评估[J].电力系统自动化，2019,43(12):18-26.GAO Kunlun,YANG Shuai,LIU Siyan,et al.Transient stability assessment for power system based on onedimensional convolutional neural network[J].Automation of Electric Power Systems,2019,43(12):18-26.
• [28]ZHAO M H,ZHONG S S,FU X Y,et al.Deep residual shrinkage networks for fault diagnosis[J].IEEE Transactions on Industrial Informatics,2020,16(7):4681-4690.
• [29]何成兵，王润泽，张霄翔.基于改进一维卷积神经网络的汽轮发电机组轴系扭振模态参数辨识[J].中国电机工程学报，2020,40（增刊1):195-203.HE Chengbing,WANG Runze,ZHANG Xiaoxiang.Identification of torsional vibration modal parameters of turbogenerator shafting based on improved one-dimensional convolutional neural network[J].Proceedings of the CSEE,2020,40(S1):195-203.
• [30]NAGI J,YAP K S,TIONG S K,et al.Nontechnical loss detection for metered customers in power utility using support vector machines[J].IEEE Transactions on Power Delivery,2010,25(2):1162-1171.