基于模糊灰色关联度的用电信息采集数据异常预警方法  

A Method for Abnormal Warning of Electricity Consumption Information Collection Data Based on Fuzzy Grey Correlation Degree

作　　者：张杨 卢霄依 张烨琳 ZHANG Yang;LU Xiaoyi;ZHANG Yelin(State Grid Shanghai Shibei Electric Power Supply Company,Shanghai 200072,China)

机构地区：[1]国网上海市北供电公司,上海200072

出　　处：《电工技术》2025年第4期105-107,111,共4页Electric Engineering

摘　　要：由于设备故障、通信问题、人为操作失误等多种因素的影响,用电信息采集数据时常会出现异常,这不仅影响了电网的正常运行,也给用户带来了诸多不便,因此提出基于模糊灰色关联度的用电信息采集数据异常预警方法。首先,收集了用电信息采集数据并提取了异常数据特征,将其应用于后续的异常模型检测中。其次,基于模糊灰色关联度构建异常检测模型,计算各指标与正常用电模式间的模糊灰色关联度。最后,实现用电信息采集数据的智能化异常预警方法。使用基于模糊灰色关联度的用电信息采集数据异常预警方法进行电力装置的测试,预警情况满足实际需求,误报率和漏报率均控制在较低水平,证明该方法能够提前发现潜在的用电数据异常,为电力公司提供足够的响应时间。Due to various factors such as equipment failures,communication issues,and human operational errors,abnormal data collection in electricity consumption information often occurs.This not only affects the normal operation of the power grid,but also brings many inconveniences to users.Therefore,a method for abnormal warning of electricity consumption information collection data based on fuzzy grey correlation degree is proposed.Firstly,the electricity consumption information collection data was collected and abnormal data features were extracted,which were applied to subsequent anomaly model detection.Secondly,an anomaly detection model is constructed based on fuzzy grey correlation degree,and the fuzzy grey correlation degree between each indicator and normal electricity consumption mode is calculated.Finally,intelligent anomaly warning for electricity consumption information collection data can be achieved.The experimental results show that after using the fuzzy grey correlation based abnormal warning method for electricity consumption information collection data for power device testing,the warning situation meets practical needs,and both false and false alarms are controlled at a low level,proving that this method can detect potential electricity consumption data anomalies in advance and provide sufficient response time for power companies.

关 键 词：采集数据 异常预警 用电信息 电力行业 模糊灰色关联度 

分 类 号：TM76[电气工程—电力系统及自动化]