| 229 | 5 | 170 |
| 下载次数 | 被引频次 | 阅读次数 |
电力系统为满足各类用户的用电需求，需掌握用户端用电负荷曲线的典型模式。当前电力负荷模式挖掘过程中往往使用单视角数据进行聚类分析，未考虑不同视角数据内部度量信息差异和复杂的计算。针对这一问题，从三个粒度获取多视角用电负荷数据并进行预处理，分别计算相似度矩阵，使用相似度网络融合算法构建融合相似度矩阵，选用Leiden社区发现算法对用户群体进行划分，然后动态识别各个子社区的典型用电负荷曲线并进行趋势分析。实验结果表明，算法能够稳定划分用户群体并识别各个子社区的典型用电负荷模式曲线，结合用户基本信息，得到各个子社区的峰值模式，从而为电力系统采取个性化措施以满足不同用户群体的需求提供指导。
Abstract：To meet the electricity demand of various customers, the power system should understand the typical pattern of the user′s power load curve. While the single-view data was often used for cluster analysis, which could not consider the difference in internal measurement and complex calculations. The multi-perspective electricity load data was collected from three granular perspectives and processed pretreatment. The similarity matrix was first calculated and the fusion similarity matrix was constructed by the similarity network fusion algorithm. The Leiden algorithm was used to divide the user groups and identify the typical power load curve of each sub-community dynamically, the trend analysis was followed. The experimental results demonstrated that the sub-community could be classified and the typical power load curve could be identified reliably by the algorithm. The peak pattern for each sub-community could be found when combined with the basic information of users and thus could provide guidance for power system to take personalized measures to meet the needs of different user groups.
[1] 陈俊艺，丁坚勇，田世明，等.基于改进快速密度峰值算法的电力负荷曲线聚类分析[J].电力系统保护与控制，2018,46(20):85-93.CHEN J Y,DING J Y,TIAN S M,et al.An improved density peaks clustering algorithm for power load profiles clustering analysis[J].Power system protection and control,2018,46(20):85-93.
[2] 方茂益，孙志杰，周国鹏，等.基于FCM配电侧典型电力日负荷曲线分类研究[J].华北电力技术，2016(11):15-19.FANG M Y,SUN Z J,ZHOU G P,et al.Typical power daily load classification based on fuzzy C-means clustering algorithm[J].North China electric power,2016(11):15-19.
[3] 朱祥和.考虑气象因素的短期负荷预测模型研究[J].数学的实践与认识，2018,48(3):131-143.ZHU X H.Study on short-term load forecasting model considering meteorological factors[J].Mathematics in practice and theory,2018,48(3):131-143.
[4] 谷紫文，李鹏，郎恂，等.基于变分模态分解和密度峰值快速搜索的电力负荷曲线可控聚类模型[J].电力系统保护与控制，2021,49(8):118-127.GU Z W,LI P,LANG X,et al.A controllable clustering model of the electrical load curve based on variational mode decomposition and fast search of the density peak[J].Power system protection and control,2021,49(8):118-127.
[5] 段秦刚，王浩浩，伍子东，等.基于改进蚁群算法的电力负荷半监督聚类[J].电力建设，2020,41(10):81-89.DUAN Q G,WANG H H,WU Z D,et al.Research on power load semi-supervised clustering based on improved ant colony algorithm[J].Electric power construction,2020,41(10):81-89.
[6] 魏震波，关翔友，刘梁豪.电网社区结构发现方法及其应用综述[J].电网技术，2020,44(7):2600-2609.WEI Z B,GUAN X Y,LIU L H.Overview of power community structure discovery algorithms and their application in power grid analysis[J].Power system technology,2020,44(7):2600-2609.
[7] 苏适，李康平，严玉廷，等.基于密度空间聚类和引力搜索算法的居民负荷用电模式分类模型[J].电力自动化设备，2018,38(1):129-136.SU S,LI K P,YAN Y T,et al.Classification model of residential power consumption mode based on DBSCAN and gravitational search algorithm[J].Electric power automation equipment,2018,38(1):129-136.
[8] 林锦波.聚类融合与深度学习在用电负荷模式识别的应用研究[D].广州：华南理工大学，2014.LIN J B.Application of recognition model for power load in cluster ensemble and deep learning[D].Guangzhou:South China University of Technology,2014.
[9] 魏伟，李永胜，程逍，等.基于多视角网络融合的典型用电负荷模式挖掘[J].沈阳工业大学学报，2022,44(3):248-254.WEI W,LI Y S,CHENG X,et al.Mining for typical power load modes based on multi-view network fusion[J].Journal of Shenyang university of technology,2022,44(3):248-254.
[10] WANG B,MEZLINI A M,DEMIR F,et al.Similarity network fusion for aggregating data types on a genomic scale[J].Nature methods,2014,11(3):333-337.
[11] CHEN J F,SUN L L,GUO C H,et al.A fusion framework to extract typical treatment patterns from electronic medical records[J].Artificial intelligence in medicine,2020,103:101782.
[12] TRAAG V A,WALTMAN L,VAN ECK N J.From Louvain to Leiden:guaranteeing well-connected communities[J].Scientific reports,2019,9:5233.
[13] NEWMAN M E J,GIRVAN M.Finding and evaluating community structure in networks[J].Physical review E,Statistical,nonlinear,and soft matter physics,2004,69(2):026113.
基本信息:
DOI：10.13705/j.issn.1671-6841.2022185
中图分类号:TM73;TP311.13
引用信息:
[1]魏伟,韩颖,刘怡君,等.基于多视角数据与社区发现的典型用电负荷模式挖掘研究[J],2023,55(04):68-74.DOI:10.13705/j.issn.1671-6841.2022185.
基金信息:
河南省哲学社会科学规划项目(2020CZH009);; 国家自然科学基金项目(72001191);; 中国科技智库青年人才计划项目(20220615ZZ07110090)