胡一伟,刘珊,黄浩.基于ARIMA和递归贝叶斯的窃电用户识别算法[J].电测与仪表,2022,59(6):196-200. Hu Yiwei,Liu Shan,Huang Hao.An Energy Theft Detection Approach based on ARIMA and Recursive Bayesian[J].Electrical Measurement & Instrumentation,2022,59(6):196-200.
基于ARIMA和递归贝叶斯的窃电用户识别算法
An Energy Theft Detection Approach based on ARIMA and Recursive Bayesian
低压窃电负荷小，难以被及时发现，给电力企业造成了巨大的经济损失。本文基于差分整合移动平均自回归模型(Auto-regressive Integrated Moving Average model, ARIMA)和递归贝叶斯算法，构建了一种针对配电网低压窃电行为的识别方法，该方法结合用户历史数据对低压用户与台区表夜间各时段电力负荷数据进行分析，并算出用户窃电概率，从而发现用户是否存在窃电行为。仿真与实际结果表明：该方法对及时准确发现窃电行为，提高配电线路线损治理效率具有重要意义。
英文摘要:
The energy theft is difficult to be discovered due to the low theft load, also causes huge economic losses to utility companies. This paper introduced an algorithm based on ARIMA and recursive Bayesian method to identify the energy theft behavior through the selected period historical data. The case study proves this approach is able to timely and accurately discover energy theft and improve the line loss management.
