基于局部离群因子的电力计量数据异常值自动化监测系统

Automatic Monitoring System for Abnormal Values in Power Measurement Data Based on Local Outlier Factors

针对电力计量数据异常导致的电力系统故障,提出基于局部离群因子的电力计量数据异常值自动化监测系统。该系统集成多传感器采集数据,通过通信模块传输至处理模块。该模块先运用AP聚类算法将数据聚类成多个类簇,再使用局部离群因子模型计算离群度,通过离群度数值得到异常类簇,则该异常类簇为异常值,再将监测结果传输到用户PC端,实现电力计量数据异常值自动化监测。实验结果表明,该系统聚类电力计量数据时的疏密度数值较高,可有效检测异常值,应用性能较为显著。

Aiming at the existing problems or faults of power system caused by the abnormality of power metering data,an automatic monitoring system for abnormal values of power metering data based on local outlier factors is studied.The system integrates data collected by multiple sensors and transmits it to the processing module through the communication module.The module firstly uses AP clustering algorithm to cluster data into multiple clusters,then uses local outlier factor model to calculate outliers,and obtains abnormal clusters through outlier values,then transmits the monitoring results to the user’s PC to realize automatic monitoring of abnormal values of electric power metering data.The experimental results show that the system has a high density value when clustering power metering data,and can effectively detect abnormal values.