基于大数据聚合的电力用户行为实时云监测方法  

Real-time Cloud Monitoring Method for Power User Behavior Based on Big Data Aggregation

作　　者：朱克 张莉 王笑一 张浩 李玮 ZHU Ke;ZHANG Li;WANG Xiao-yi;ZHANG Hao;LI Wei(Marketing Department of State Grid Corporation of China,Beijing 100031,China;State Grid Customer Service Centre,Tianjin 300300,China;Beijing China-Power Information Technology Co.,Ltd.,Beijing 100031,China)

机构地区：[1]国家电网有限公司营销部,北京100031 [2]国家电网有限公司客户服务中心,天津300300 [3]北京中电普华信息技术有限公司,北京100031

出　　处：《计算技术与自动化》2022年第4期173-178,共6页Computing Technology and Automation

摘　　要：为提升电力用户行为监测效果及准确性,判断电力用户异常行为,提出一种基于大数据聚合的电力用户行为实时云监测方法。该方法将基础设施及终端等获取的电力用户行为大数据储存至数据层的关系数据库内,处理层调用数据层存储电力用户行为大数据,采用大数据处理技术,通过数据降维、清洗以及标准化处理后,提升电力用户行为大数据质量;应用层采用改进流数据聚类算法,通过用户及簇典型曲线提取、曲线相似度度量,实现用户用电行为异常监测,并通过显示层云展现监测结果。实验结果证明,该方法的数据聚类质量高,可以有效获取电力用户行为监测结果,判断电力用户是否存在异常行为,具备较高监测准确性。In order to improve the effect and accuracy of power user behavior monitoring and judge the abnormal behavior of power users,a real-time cloud monitoring method of power user behavior based on big data aggregation is proposed.The big data of power user behavior obtained by infrastructure and terminals are stored in the relational database of the data layer.The processing layer calls the data layer to store the big data of power user behavior.The big data processing technology is adopted to improve the quality of big data of power user behavior through data dimensionality reduction,cleaning and standardization;The application layer adopts the improved stream data clustering algorithm,realizes the abnormal monitoring of users’ power consumption behavior through the extraction of typical curves of users and clusters and the measurement of curve similarity,and displays the monitoring results through the display of layer cloud.The experimental results show that this method has high data clustering quality,can effectively obtain the behavior monitoring results of power users,judge whether there is abnormal behavior of power users,and has high monitoring accuracy.

关 键 词：大数据聚合 电力用户行为 实时云监测 M-BIRCH算法 相似度度量 典型曲线 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]