基于局部异常点检测算法的电能表飞走异常智能分析

摘要

针对用电信息采集系统在计量设备在线监控业务应用的实际情况，文中介绍了一种基于LOF局部异常点检测算法的电能表飞走异常分析方法，该方法可以通过计算电能表示值异常因子，有效判断电能表示值异常属于偶然异常还是趋势性异常，在排除偶然数据的影响后，进行电能表飞走异常的二次判断。该方法可以有效提高该类异常判断的准确性，避免故障误判，提高了现场故障处理的工作效率。%Aiming at the actual situation of the on-line monitoring and controlling system of the electric information ac-quisition system , this paper introduces a method of detecting the abnormal point of the electric energy meter based on the LOF( loacl outlier factor ) detection algorithm , which can be used to detect the occasional anomaly or the trend a-nomaly , and conducts the secondary judgment of trend anomaly after eliminating the influence of accidental data .This method can effectively improve the accuracy of this kind of anomaly judgment , avoid the misjudgment of the fault , and improve the working efficiency of the on-site fault treatment .