基于灰色分析的集抄数据异常判定

Data Anomaly Determination for Centralized Metering Systems Based on the Grey Analysis Methods

提出综合缺陷分析和失真分析的集抄数据异常判定流程,着重针对目前用电量环比波动率判据的不合理性,运用灰色分析理论构建集抄数据失真判定的两阶段法:第一阶段通过电量折算削弱季节变迁对用电环比波动率的影响;第二阶段实施灰色关联度分析判定用电模式稳定性,对用电模式稳定的用户采用灰色距离测度法估计月用电量合理区间,以此作为集抄电量异常判据。对上海市北地区8 643个计量点2015年7月的集抄数据分析,结果表明所提方法能降低气候变化或特殊用电模式造成的集抄数据异常误判,合理降低复核工作量、提高出账率;所提方法还可同时找出窃电可疑用户、一定程度上克服集抄环境下的窃电隐蔽问题。

ABSTRACT：A flowchart for data anomaly determination is presented in this paper,which is able to identify both defects and distortions in the centralized metering data. In the light of the irrationalities of the currently used consumption chain volatility criteria,a two-stage method is especially proposed based on the grey analysis theory for metering data distortion determination. In the first stage,impact of seasonal temperature changes on the consumption chain volatilities is weakened by converting consumption to a unified temperature condition. In the secondary stage,consumption mode stability is analyzed by the grey relational analysis method. For customers with stable consumption modes,rational monthly consumption ranges are evaluated by the grey distance measure method and used as refined criteria of data anomaly. Illustrative example is presen-ted based on the metering data of 8643 low-voltage customers in the north area of Shanghai in July 2015. Results show that the proposed method can reduce misjudgment in metering data anomaly analysis caused by temperature variation or special consumption modes,reducing the burden of artificial metering review and increasing accounting rate of the centralized meter-ing system. Apart from this, the proposed method can also identify electricity stealing suspects so as to alleviate to a cer-tain extent the concealment of electricity stealing in the environment of centralized metering.