基于配电物联网的反窃电预警系统研究及应用

基于电力物联网建设方向,结合大数据、人工智能深度机器学习技术,提出了基于配电物联网的反窃电预警系统研究及应用。通过配电台区"变-线-相-户"分段、分层的窃电台区嫌疑分析,客户用电负荷曲线的特征分析,精准锁定窃电嫌疑用户,提高供电单位窃电预警能力。首先,采用边缘物联代理技术采集用电特征数据,获取电表的电气数据及自动拓扑关系;其次,采用聚类、分层分析建立反窃电预警模型并结合专家诊断库生成窃电嫌疑用户清单;再其次,警电联动应用将窃电嫌疑用户推送至公安侦办系统,形成警电联动体系,最后展望系统扩展对相关业务的支撑,旨在探讨电网企业应用反窃电预警系统的价值。

Research and Application of Anti-stealing Early Warning System Based on Distribution Internet of Things

Based on power iot construction direction,combined with large data depth,artificial intelligence,machine learning technology,put forward the power research and application of the early-warning system based on the distribution of the Internet of things。Through the distribution area"Transformer-line-phase-door"segmented,layered power suspected area analysis,customer characteristic analysis of electricity load curve,precision lock suspected power users,improve the capability of power supply unit power theft warning.Firstly,the edge coupling agent technology is used to collect electricity characteristic data and obtain the electrical data and automatic topological relationship of the electricity meter.Secondly,cluster analysis and stratification analysis are used to establish the early warning model of anti-electricity theft,and the list of suspected users of electricity theft is generated by combining the expert diagnosis database.Secondly,the combined application of alarm and electricity pushes suspected users to the public security investigation system to form a combined system of alarm and electricity.Finally,it looks forward to the support of the system expansion to related businesses,aiming to explore the value of the application of anti-alarm system for power grid enterprises.